In many respects the office has changed little over the past 200 years. The introduction of the telephone, copier, fax and computer have only served to speed up and proliferate the basic processes. We are now faced with an increasingly complex and a difficult environment that requires fundamental changes to humanise the processes. In this paper we address some of the interface issues that now appear to have near term solutions.

PROLOGUE
Life has evolved over the last 600 million years with man coming to true superiority in the last 100 thousand or so. This evolutionary process has equipped us for an environment consisting of modest size entities moving at reasonable speed in reasonable time frames. In contrast the evolution of technology during the past 60 years has placed us in a foreign environment with massive or minute entities moving at incredible speed in very short time frames with very little or no perceptive delay between events. That is: the electronic evolution has now outstripped the rate of our biological evolution to cope with change. Moreover, our interface with technology has generally been designed for the convenience of the technology and is not intuitive or biologically matched to our abilities. If we are to change the office and the working environment significantly in the future, then these issues have to be addressed and the interface has to have a human orientatation.

Computer and communications technologies now look ripe to introduce some radical and long overdue change. All the technology and know-how is available (in abundance) to revolutionise the office, the home and the place of work far beyond the evident results of incrementalism we currently enjoy. In many respects we might now therefore consider the modern office to be an unnatural, and even hostile environment for most humans. We currently suffer interfaces and conditions that are not convenient, user friendly, or conducive to efficient and pleasant operation. This applies for the interfaces between humans or individuals and machines. This is not however, in our view, a necessary condition and perhaps more importantly, it is not sustainable in the long term. The question is: what is sustainable, and what happens next?

Addressing this question and the likely solutions to the proceeding problems presents the main theme of this paper. We thus explore a series of proposals for potential future office environments. These follow an underlying holistic approach to the integration of existing and new technologies to create a new IT environment. This is integrated, intuitive and responsive to human needs. It also places both the user and the tasks/work as the central focus.

WISHES
I wish:

• My phone was always with me
• My desk wasn't covered with machines and cables.
• I could find people when I want them, and they could find me.
• I could get rid of all this paper.
• I didn't have to travel so much
• I could be in two places at once.
• My PC wasn't so thick and unfriendly
• I could simultaneously see multiple (and full) electronic pages.
• My PC screen was flat, horizontal and had the qualities of paper.
• I could talk to my PC and it to me.
• My diary and daily schedule was automatic and up-to-date.
• I had an automatic database of all my contacts.
• My mail was electronically sorted, summarised, prioritised & filed.
• Information could be retrieved with only a partial description.
• I could voice annotate documents.
• I had all the power of my office wherever I was.
• It didn't take 3 months to move office and restore full IT.
• Etc, etc.

• Limited office wiring and communication bandwidth.
• Constraints of cordlessness and portability.
• Teleconferencing was not so limited and lacking in realism/facilities.
• Too much THICK (un-intelligent) paper.
• Information overload, categorisation, filing and retrieval.
• Inadequate and in-human interfaces.
• Multiple devices which don't easily integrate.
• Storage and processing power available only as hardware.
• Etc, etc.

Specific features of the desk include: optical communication that is cordless and large bandwidth; built-in equipment and an active surface for document display, manipulation and cordless/active peripherals; multi-standard input and output devices; intelligent non-intrusive interfaces; software filing, summarising, and correlating; intuitive and ergonomic control systems; built in recognisers for 'hot desking', with a secure data environment; teleconferencing with human scale interactive images; hi-fi acoustics; voice I/O and command. Let us look at some of these features in detail in the sections that follow.

OFFICE WIRING
One of the major limitations of present day office design and realisation is the necessity for hard wired desks. Even with the exciting optical fibre technology developments there still remains an underlying problem with the cabled office: getting fibre or cable to where you want it. With the increasing demands for more communication this is likely to become even more problematic in the future. In most sectors the speed of market and technology induced change implies regular re-organisations and movement of staff and operations. It is not uncommon to find that moving office currently involves a delay of some 3 months before all electronic back-up systems are terminated at your desk.

Optical wireless affords an important means of short range, diffuse and line-of-sight fixed and mobile communication for inside the office without the regulatory or frequency restrictions of radio alternatives. Furthermore the bandwidth of the channel is potentially as broad as cable based optical fibre systems, thereby allowing broadband multi-channel services. The principle is directly analogous to radio. Data can be omnidirectionally radiated from a ceiling, desk or body mounted antenna and transceiver (Fig 2) Transceivers using holographic dispersers can illuminate very well defined cells so that different data domains can be accurately positioned and addressed within the office environment (Fig 3).

So with optical wireless the office can have an omnipresent optical ether so that people and their desks can be mobile and still have broadband communication. People and equipment are thus free to roam within a building with no more data, printer, fax or telephone cables - only power is required.

The optical ether also enables the use of a lightweight headset with microphone and earpiece to provide cordless communication (Fig 4). Furthermore, voice recognition software allows direct voice I/O with computer and communication systems. With intelligence built into the cellular optical wireless system the headset can be tracked and automatic location and activity systems can be used to produce "who, where and when" activity databases. Combining voice recognition and the location facility provides a secure method of "hot desk" operation anywhere in an office. Talk to any desk and it can check your identity and configure to your own personal definition using the broadband optical communication to access your virtual desk's facilities.

THE DESK
Today desks are passive objects on which we stack, and in which we store, things. Technology has made them a mass of wires, equipment boxes, keyboards, mice and phones; none of which easily work with each other and all with their own proprietary interfaces. The wiring alone causes configuration nightmares whilst the integration of diverse software and hardware is rapidly approaching the impossible. One solution to this is an active desk with: a built in optical backplane; a partitioned structure used to house equipment; an inductive working surface to provide battery charging and communication to cordless peripherals; ergonomically built-in multiuse displays and input devices; radically new user interface such as "hands in the screen(Fig 5)and eye plus voice tracking".

If optical wireless is used in the office, then cordless objects could be used instead of wired mice and keyboards utilising the optical ether. Inductive loops printed below the surface of the desk (like a car's heated rear window) would charge anything placed on its surface. A laptop or active organiser placed on the desk would be trickle charged at the same time as communicating with the desk allowing the full processing power of the desk to be instantly available without any physical connections.

With equipment slots built into the structure of the desk multi-vendor devices could be added (like shelves in a racking system) and communication links established. This would mean a device such as a CD WORM unit could be purchased and just dropped into a slot. Intelligence in the desk would establish power and communication and integrate it into the computer workspace as well as displaying its controls on any desired display.]

VIDEO CONFERENCING
Video conferencing has the ability to radically reduce the need for people to travel and can also deliver a new team working media for geographically dispersed organisations. With the addition of telepresence hardware, a person can literally "be in two places at once". The constrained bandwidth available today for this human interface currently produces visual anomalies in the perception of the images and is detrimental to realising its full potential. To improve and humanise the limiting aspects of videoconferencing a different type of interface is proposed.

A large rear projected HDTV monitor can be ergonomically placed (Fig 6) in the desk. This produces high definition life-size images in front of the user (in a natural face-to-face mode). By the use of an LCD shutter as the screen material a video camera can be aligned to be looking directly at the user through the screen. This enables a human sized image of your conversant with eye to eye contact and gaze awareness. Because of the large size of display the peripheral vision would be partially filled and create a feeling of " being there" rather than watching a picture.

As this display is a High Definition, then it can also be used as a computer monitor and in many applications allows the mixing of videoconferencing and computer generated data.

By using an infra red emitting pen the screen can also be turned into an electronic whiteboard (Fig 7) via infrared sensing in the camera driving the cursor controls of the computer. This allows multiple videoconferencing participants to work together in the same electronic media space in real time. People sitting at desks thousands of miles apart come together in an electronic media to realise real time team working.

HANDS IN THE SCREEN
The preceding user interface has been realised by an integration of currently available technologies and working practices. However, the addition of an overhead camera, scanning the desk's surface, and producing a positional image of the user's hand ("or finger worn" 3D RF positioning sensors) allows the realisation of an economic "hands-in-the-screen" interface (Fig 5). This direct hand control and manipulation of objects is linked to the function of the computer and peripheral equipment. No keyboard or mouse control is necessary; just speak the text and then "grab it" and put it where you want it.

Current work is centered on 3D data visualisation. A "hands in screen interface" allows modelling and manipulation of data and virtual objects. These can be placed in the medium viewed through the screen and directed by a combination of voice and hands. To further enhance the lifelike and intuitive nature of this interface, 3D technology is being introduced to add a depth of vision, dimension, reality, and personality to the environment. Objects are being humanised to react emotionally and give heuristic guidance during interactive sessions with movement, stance, colour and/or audio to convey reactions. For example: icons try to avoid your hand if the action is questionable, or become defensive if you are about to initiate a damaging direction of actions. Alternatively, you could move your hand and grab a document, pull it to the print area and it would wriggle and complain as you hadn't yet reviewed the spelling but it wouldn't actually prevent you from printing.

ELECTRONIC POST-IT
To ensure that the main working display is not crowded with buttons, icons and electronic messages, another simple display with a touch sensitive surface and voice activation can be appropriately positioned. This can be used for telephone directory listings, "post it" pads and soft keys for all desk controls. For example this enables an up to date electronic directory to be displayed and a telephone call established whilst still being part of a video team working session. This also acts as the control panel for all equipment that is installed in the desk - no more front panels on boxes you can't reach.

To aid use, frequently used functions remain large and easily activated while less frequently used controls gradually migrate to a lower control layers. To further enhance the intuitive nature of these controls they can also be assigned "emotion". For example, you might say, " Phone Granny". The display would show both grandmothers' names and a dot would impatiently dither between the two. A straight "Granny Fisher" statement then prompts the phone dials out.

PAPER
The paperless office does not exist and might be perceived to be less likely with time as more technology is introduced that demands more copies. This trajectory is non-sustainable in the long term and we have to reverse the trend and halt the growth in paper proliferation.

The user interface to electronic mail systems can be radically improved if our human-oriented user interface is applied with a few minor enhancements. For example; the scanning of bar coded documents allows automatic logging, filing, abstraction and tracking. For example, the document arrived at 9:15 am on 27th October. Rob and Phil were with you plus a visitor. The text correlator reads the central file copy and the key words are 'Information Exchange' and 'publication date'. This related information, when automatically appended, enables single location filing and retrieval via sparse descriptors. This falls precisely in line with our abilities. As humans, we can vaguely remember the scenario: "Rob was with a visitor and it was in the morning". All the documents in this category, complete with a video snap of the visitor, can thus be recalled.

As we move to a multimedia environment then the ability to add colour, moving images, sound and interaction to documents will lead to paper being a less powerful medium. Electronic mail will then include video sequences, active directories and databases in a form that match your desk's personal 'sifter' and organiser.

MEMORY, PROCESSING & COMMUNICATION
These facilities and anticipated amounts of data are going to place heavy demands on memory and processing power. However, with the cost of both dropping by an order of magnitude over the last 5 years, and the trend looking set to continue, then the technology should be within the reach of most offices within the next decade.

In order to reduce the memory required, a process of (Hebbian) data decay is being investigated. Documents are reduced in data content with time as their perceived importance diminishes. Thus a document with full colour and voice annotation decays with time through to a monochrome document with low quality audio. Finally it is compressed with only contextual and retrieval information easily accessible. Regularly used or vitally important documents can remain uncompressed and complete.

FINAL REMARKS
All of the technology described is either available or currently under development. However, very little has yet been integrated into a complete system that in any way reflects the potential gains possible. Creating an environment in which people are able to work intuitively, organise information and interactions on a human scale should be a prime objective. The proposals and initial investigations presented here are a first step towards that intuitive "Office I wish I had". Our aim is to break each of the interface barriers by a human orientation of technology to release the joint intellect of man and machine. The continued exponential cost reduction in electronic data storage, processing power and communication bandwidth now make this a real possibility. A decade from now could see it generally available in the work place.

BIBLIOGRAPHY
1] Kruger, M W: Artificial Reality II, Addison Wesley, 1991.
2] Cochrane, P et al: CAMNET, Interlink 2000, Aug 1992.
3] Ishii, H & Kobayashi, M: ClearBoard, CHI-92 Conference May 1992.
4] Cochrane, P & Heatley, D J: Optical Fibre Systems & Networks, ibid [2] Feb 92.
Fig 1 Future Desk
Fig 2 Optical wireless in the office - ceiling satellite
Fig 3 Holographic lensing and micro-cellular
Fig 4 Optical headset
Fig 5 Hands in the screen
Fig 6 HDTV Back projection
Fig 7 Electronic Whiteboard