New Scientist Interview of Thad Starner

by Peter Thomas

You emphasise that 'cyborgs' don't really require direct neural interfaces of the kind commonplace in future-shock fiction (Gibson, for example). Surely this is the endpoint for 'wearable' computing, and some of the augmentations you suggest are intermediate technologies. Will wearble computing move on to direct neural interfacing, if so on what timescales, and what are the immediate barriers?

Neural prostheses are interesting, but I don't see an application in wearable computing, except for the disabled. The most obvious suggestion for a neural prosthetic is an improved display interface. For some applications, my current display (the Private Eye available from the Phoenix Group) is too obtrusive. Thus, many people feel that going directly to the visual cortex (back of the head) with electrodes is the answer. Unfortunately, this does not work too well. First, any electrode that is pushed into neural tissue will cause the immediate tissue to die over time. However, it is possible to sever nerves and allow them to regenerate through electrode rings. This may provide a stable connection. The problem is finding the right nerve and hoping it regenerates properly. Needlesstosay this is difficult in a area as densely packed as the visual cortex. There is yet another obstacle to the visual cortex interface. Stimulation in this area does not lead to a one to one mapping in the visual field when trying to display more than one spot at a time. This is because 5 layers of visual processing are already performed at the retina. In fact, looking into someone's eyes is the easiest way to see their brain! Thus, until we understand visual phsyiology better, the best place for a prosthetic display is in the eye. However, the retina is very fragile and interfacing to it has not, as yet, proved fruitful. A more promising idea is to make a display that can be used like a contact lens or mounted inside the eye without touching the retina. Phil Alvelda's work on microdisplays may make this method possible (web address). Also, non-intrusive displays are already being made much smaller and unobtrusive. Therefore, the market for neuroprosthetic displays seem limited to those people who can see no other way.

Aural neuroprostheses already exist. However, these are again limited to medical conditions. Pushing electrodes into the inner ear tends to destroy the normal structures for hearing. The best case for these cochlear implants, to my knowledge, is a five electrode device which is generally billed as an assistant to lip reading. For those with normal hearing, small bone conduction earphones/microphones are the obvious choice.

Having covered sound and vision, the remaining senses are touch, taste, and smell. Smell and taste we are just beginning to understand, and I am not qualified to talk about prosthetics for these senses. However, touch has some potential. These nerve bundles are more accessible. However, again, less intrusive interface mechanisms may limit the market for such devices. The proprioceptive system can be fooled by muscle vibration, and some of the VR glove equipment has showed that piezoelectic elements can provide a sort of tactile feedback.

A much better case for neuroprosthetics is for computer input. Research in bionics show that muscle activation signals can be used to control prosthetic arms, and EEGs can be used, albeit slowly, for input as well. Thus, it is easy to imagine a typing method using neuro-electical signals. Furthermore, neuro-devices for sensing the mood of the user (and having the computer adapt appropriately) may also be useful. However, the encumbering wires of EEG's and the reluctance to cut into functioning tissue by the medical community will probably limit the availability of these sorts of neuroprosthetics for a good time to come.

More sophisticated devices, such as augmenting the brain directly with computer memory, seem decades away at least. We simply do not understand enough of the brain for such an undertaking.

The question was "will wearable computing move to neuroprosthetics and what will the barriers be?" First, I think a clear need for the invasive technology has to be demonstrated. At this stage I do not see the need except for the disabled. Secondly, a finely controllable, long-term method for interfacing to neural bundles will have to be invented. This will take many years. Finally, the reluctance of the medical community to operate except in cases of illness and cosmetics will have to be overcome. This may prove a harder step than most people realize.

In 'human powered wearable computing' you describe a number of power alternatives. Which of these have been realised, to what degree, and how successful have they been?

We are just beginning to experiment with the different power generating mechanisms discussed in the paper. The body produces at least 81W at any given time and over 1 kilowatt maximum. If only a fraction of this can be harnessed without encumbering the user, a wearable computer can run without ever needing an electrical socket.

The most promising method from the paper is the power shoes. The concept is to both improve the efficiency of the user's gait while simultaneously generating up to 17W of power for the computer while at a brisk walk. We are trying to prototype such a system now. Another method involves making clothing out of piezoelectric film. Every movement would generate charge. Apparently such a device was made in the mid 1980's, but I still haven't been able to track down the research. Other methods discussed are best reserved for the medical community or niche applications.

Some of the power alternatives you discuss in fact do have 'cyborg' overtones of the type common in future-shock fiction. How do you think these ideas would be received if they ever appear as commercial products?

One could say power shoes are a commercial product. Certain sneaker manufacturers already put flashing LED's in the heels of the shoe. Power generation for wearable computing is not far away from that technology. In addition to generating power, the shoes can also contain the computer. Interface devices can generate their own power from the user's motions. For example typing produces on the order of 10 mW of power (thanks to Prof. Hawley for suggesting this). As a last resort, some devices could use small batteries, but these batteries would need much less maintainance since the main power requirement, the cpu, is not draining them.

To connect all these different devices we hope to use a network that harnesses the body's natural electrical fields for communication. This network, developed by Tom Zimmerman here at the MIT Media Lab, requires very little power to operate and is difficult to tap unless the person is in physical contact with the detector. Thus, a wireless, relatively self-maintaining wearable computer may be possible.

Many of the possibilities for wearable computing power seem to be targeted on ubuiquitous communications. Do you think that completely ubiquitous communications (the cellular communicator in the ear, for example) will become pervasive, and on what timescale?

No, not in the bandwidth desired. Having said that, I hope I'm proven wrong. However, for example, this file was typed walking to a cab in Boston, boarding a plane, transferring in Washington DC, and flying to Switzerland for a conference. Currently I'm typing in a hotel lobby waiting for my room to be ready. At several points in this journey wireless communication would not be possible, due to bad design and paranoia by the airline industry for their navigation systems, but also due to technical problems of providing wireless world coverage. While satellite systems can provide a low bandwidth link, the bandwidth necessary to handle a large consumer base running multimedia applications is daunting, especially at current allowable emission levels. Thus, I believe wearable computers will always need local cpu and storage to provide a consistent interface.

On what are your definitions of 'persistence' and 'consistency' based for 'real personal compurters'

Persistence simply means that the computer is always with you, always on, and always accessible. Note that while pen computers can meet the first two requirements, they do not meet the last. In order to use a pen computer, it must be unpackaged, turned on (if it isn't already), the stylus must be found, attention must be focused on the screen, and then both hands must be used to write. Thus, it would be impossible to use such a system at a conference. Compare this to a wearable computer with a heads-up display and a one-handed keyboard. With my wearable I can store names and interesting snippets of conversations while shaking hands and maintaining eye contact during professional meetings. I have easy access to my computer while walking the halls of MIT, swinging a rock hammer, or fighting my way through a blizzard. While I can't claim to wear my computer all the time, that is a goal. At conferences, like the one I'm currently attending, I wear my computer almost every waking hour to help organize the onslaught of information that occurs.

Consistency means that I always have the same interface available to me no matter what the computing environment. For example, currently most of my computer interactions are managed by my wearable computer. If I start to use a DEC Alpha workstation, it should automatically accept input from my chording keyboard and return results, if desired, on my heads-up display. Bank machines, subway fare takers, and racetrack betting machines should interface similarly. Not only would such a system be more hygenic (public keyboards are a good carrier of germs) but would also provide a more convenient interface for the user. For an extreme example, the visually disabled could use custom displays to help them access the bank machines the sighted take for granted. On the other extreme, in richer environments where gesture and speech are used such as our Interactive Video Environment, the wearable can act as a base level private interface as well as an arbitrator for the advanced features the more powerful computers provide. Since so much time is spent with the wearable interface, users tend to get very proficient with it and customize it to their needs. Indeed it is worth their effort to customize the interfaces since these devices (at least the ones I'm interested in) are designed for long term, intimate relationships.

One argument runs that future intimate computers are only extensions of current prostheses of various kinds. Are in fact intimate and wearables of the same order, what are some of the key differences, and under what circumstances might a marketing campaign for commercial products be successful.

In my research, wearables and intimate computers are the same. I tend to use the term "wearable" due to the misconnotations the term "intimate" computing seems to bring. However, I think wearable computing is the more inclusive term, including those machines that are, or will be used for inventory control, stock brokering, military command and control, security, public speaking, newspaper distribution, insurance adjustment, policing, real estate, athletics, and medicine, just to name a few. I personally do not tend to include pen systems in this classification because they are not easy to access in a wearable fashion (with one or two exceptions), but that is my personal bias.

Wearable computers are/can be immediately successful in several niche markets. For example, the U.S. Army has already field tested augmented soldiers and is spending a considerable amount of effort to further their usefulness. Other near term markets center around those who need real-time information. For example, brokerage firms need to react quickly to changing markets 24 hours a day. Here wearables are ideal, since the broker can be instantly alerted to a market change and react while still walking back from lunch. News correspondants can also use the reaction speed of wearable computing. Imagine the chaos when interviewers can instantly check a politician's statement against previous promises! Computer system administrators can immediately benefit in that they can monitor their systems for intruders or perform maintainance anywhere, anytime. There are several other initial markets, and once wearable computers start to move into a field, sales may become exponential as businesses rush to stay competitive. In the future, I think wearable computers will become a status symbol like beepers, cellular phones, or, for that matter, a good suit. In fact, there are already efforts to invoke fashion with wearable computing.

As the equipment gets less obtrusive and more powerful wearable computing will turn mainstream. Almost every college student will want one for notetaking, especially when the new computer savvy generation comes through. With the augmented memory software we're building, students will be able utilize more of their education than ever before. Medicine is also an obvious market. Next time you visit your doctor take careful note of how he handles information. In general, doctors do not write in front of their patients during an exam. Apparently, the idea is that the doctor does not want to influence his patient's answers or scare him with diagnoses. However, according to the doctors, this also means that information is lost between the time of the interview/exam and when it is recorded. Unobtrusive wearable computing would allow instant private recording of information and real-time querying of large medical databases for precedents. This corresponds to better care for the patient. Imagine the help this would have been for the discovery and diagnosis of AIDS or the containment of even faster spreading epidemics!

An initial way to penetrate this market and overcome reluctance is through liability insurance. Apparently one of the major costs for the insurers is organizing the medical paper trail to try to defend doctors. Unfortunately, the doctors' own handwriting is working against them. However, by giving doctors discounts on their liability insurance if they produce their notes on a wearable, this cost of organizing patient history information is minimized.

Just like automobiles, paging, and cellular phones, wearable computing will be focused at first on rich specialty markets that have insatiable needs for real-time information. Then, as the general population gets exposed to these devices and production lines get established, larger markets will come into play and users will find their own applications. This will have tremendous repercussions in the hardware, software, communications, and service industries.

Could you outline some of the key developments (in comms. s/w h/w, circuitry, battery technology)which will impact on wearable computing until the millenium?

A five year outlook. Hmm. The number one development will be unobtrusive displays. Numerous prototypes already exist. Miniscule low power computer systems will also improve the attractiveness of wearables. For example, both OKI and SMOS have full 486 PC credit card sized computers on the market. Unfortunately there is still an issue of connectors with these systems, so maybe I should list high quality, high pin count, small scale connectors as a key development! PCMCIA is a good starting point, but small details like the size of a VGA monitor connector can mean the difference in a consumer's mind.

Network standards are of prime importance. Not only is this important for communications from the user to the outside world but also communications among the different devices on the user's body. For example, there is an amazing amount of hardware redundance between the pager, cellular phone, notebook computer, and PDA's. Why not have one display device, one communications box, and one user input manager which all these different devices can share through a body centered network? The consumer should be able to buy such a module off the shelf from any vendor, put it in his pocket, and have it automatically work with his other devices. Of course, a set of standards are needed for such a paradigm to work.

Finally, given that the hardware is coming, what will be the software developments that will make wearable computing attractive? Beyond the specialty niche applications lie two domains that I think are especially important: augmented memory and augmented reality.

Augmented memory is probably the "killer application" that will sell wearable computers to the masses. Imagine having a system that continuously "listens" for the context of the user and automatically suggests useful information. For example, we have a face recognizer that runs on wearable-class machines. When greeting a colleague, your "Remembrance Agent" (RA) notes with whom you are talking, the time and date, and the place through the Global Positioning System. Using this information and any notes that are typed during the conversation, the Remembrance Agent unintrusively suggests the top five pieces of information it thinks might be relevant to the conversation. This information is continuously updated as the Remembrance Agent gains more knowledge of the situation. Useful suggestions might include the colleague's most recent piece of correspondance, his contact information, a professional paper, a file on a related topic, or simply the fact that he owes you money. Other benefits, besides timely information, include organization, sharing of knowledge, prioritization of information, and continuous "brain-storming."

Organization is provided automatically in that the RA provides potential files for storing new information. Knowledge can be quickly shared by trusted colleagues by simply allowing the RA to automatically respond to queries for information by returning the appropriate file. Thus, instead of interrupting work with a phone call, queries for information could be sent via e-mail to the person's RA. When a general community uses Remembrance Agents, questions can be electronically broadcast to the community, and the questioner's personal Remembrance Agent can prioritize the replies. Common sense takes on a new meaning! The RA can also be modified to prioritize incoming information such as e-mail. Its like having an adaptive, very personal secretary. Finally, an unexpected side benefit is the associations between ideas the RA generates. The first time I ran the prototype RA, a new research project was born from the previously unseen connection it made between two ideas. Thus, the Remembrance Agent can act as an assistant to creative thinking.

The ability of wearable computers to provide views of both the real and virtual worlds at the same time makes them ideal for augmented reality. Imagine a searchable phone directory overlaid on the physical telephone, a virtual index overlaid on a file cabinet, 3d instructions overlaid on the photocopier machine showing how to clear a paper jam, or a "post-it" movie reminder to put out the garbage. Imagine the whole computer "desktop metaphor" extended to reality or virtual museum tour guides or illustrating a point by drawing electonic diagrams in space with a finger! The power of this paradigm is just beginning to be explored.

In summary, the key developments in wearable computing will first be hardware that makes these computers accessible to the common business worker. This will be followed by a new software revolution that will make computers take on the task of a personal assistant as never before.

Last modified: Tue Sep 12 14:55:44 1995