Go to Xanadu for a description of the architectural concepts


In Xanadu did Kubla Khan
A stately pleasure-dome decree:

Where Alph, the sacred river, ran
Through caverns measureless to man
Down to a sunless sea.
So twice five miles of fertile ground
With walls and towers were girdled round:
And there were gardens bright with sinuous rills, Where blossomed many an incense-bearing tree;
And here were forests ancient as the hills, Enfolding sunny spots of greenery.

Samuel Taylor Coleridge - 1797


The Xanadu Project


2004 to 2000
With the HP Star Team • KnOwhere

Go to “Post Action” for feedback to this page

Design Concepts by:
Matt Taylor

• HP Eco-BuildersDesign Work & Star Team/MG Taylor
DesignShop experience May 2000 •


This site describes the development of a series of environment concepts for the Open Innovation Park. The purpose is to give expression, starting with prior concepts from the last 20 plus years, to the various specific ideas being developed for the Innovation Park. These will document one possible THERE to HERE path.

It is possible that some of these architectural concepts may survive and become built. New ones may be developed as work progresses. More importantly, however, is that they contribute to a the creation of a BRAND concept and image - a unified architectural theme that can permeate each environment that the OIP actually employs.

This page will focus on two aspects of the opportunity. First, the Trade Dress and Brand issues as they relate to Open Innovation Park environments and, second, the “THERE to HERE” steps associated with creating a series of environments that provide increasing functionality (through a series of iterations) while maintaining and developing a consistent Brand essence throughout the process.

An explaination of the basic architectural concepts and features of the “End State” concept can be found eleswhere. The programmatic and process related aspects of the design will be covered here.


The physical layout of Xanadu work areas is based on the Open Innovation Park concept of the HP Star Team which was further developed in the May DesignShop experience. Go to the event Documentation for details. In essence, as users develop their ideas they travel through different zones that are arranged, equipped and staffed for that specific stage of their process. Four of the five outer Domes are set up for this function. This is compatible with Patches in the PatchWorks Design process. The layout, therefore, is a metaphore of a global work experience and process.

An early expression of this kind of layout schema was the proposed but not built Vanguard NavCenter 1995 which had clusters of “Power Teams” doing rapid prototyping surrounded by support facilities of all kinds.

The large hexagonal platform, indicated on the Plan View and shown on the Section, is dedicated to client/partner “incubation” areas. There are other platforms in the various geodesic spheres not shown on these drawings (for clarity and scale reasons).

The larger center dome houses the “KnowZone” the main circulation paths and other commons. The large platform (in place of where the 6th dome would be) is an elevated indoor/outdoor area that serves a variety of recreational, entertainment and logistical functions.

Xanadu also has extensive live in facilities in the three Towers that rise above the domes and in mini-towers that are scattered about the site. This aspect of the program derives from the “Knowledge Factory” concept of 1986.



The “End State” Model embraces a full venue 24/7/365 operational model. A complete environment for creativity. Specifically, the Park directly supports the innovation process through prototype and business planning and formation.

The “End State” environment is populated by ValueWebs whose various members come to “play” the innovation game every day. Only a very few of the total population are OIP staff who function in the “System Keeper/Integrator” and KnowledgeWorker roles. All support services, for example, are performed by ValueWeb members according to the performance specifications provided by the OIP Team. This is true for hotel and food services, for design expertise, for physical prototyping, as a few examples.

Therefore, the exact configuration of capacities and people change on a daily, weekly and monthly basis as is required by the work. Users pay a daily cost to come - this pays the overhead. The Park then participates (in some way) in the outcome of the work.

At the End State, the word “Park” is not a metaphor but an exact description of the environment. A Park of innovation with all the components there necessary to test and prove out concepts and to create a real organization.

Some of the satellite buildings on the acreage may be built and maintained by ValueWeb members who offer specialized services to users in the process.



Exuberance, abundance, creativity, playfulness, other-worldly-ness, unlimited opportunity, sensualness, natural setting, recreation, renewal, exploration, cityscape, technological (and seamless) wizardry, integrity, extreme sense of shelter, and deliberate ceremony and ritual make up the major theme elements of this design. While being an environment that is deeply integrated with nature, it must strongly execute the deliberate thematic elements of being an innovative and artistically executed human artifact. This is integral to the purpose of the structure which is to house the innovation process.

These themes are reinforced, at every opportunity, by use of denotation and connotation in the design, by the stressing of specific programmatic elements and by the “extreme” viewpoint expressed by the structure. This later aspect is easy with the fully developed version shown. It becomes more difficult in the earlier, more simple and modestly scaled works that can easily default to a common or conventional viewpoint. This is a critical thematic element, however, and each work - no matter where on the iterative development cycle - must express a clear, unique, idiosyncratic view of the world. Innovation comes from such a attitude. This must be done while, at the same time, accomplishing Pattern Language values and employing solid architectural principles. In other words, the environment must not be merely “showy” or eccentric.

These thematic elements must be handled in such a way that two usually opposite experiences are integrated and predominate for the users: excitement and serenity. Excitement is roughly related to the use of Prospect; serenity to Refuge.

Denotation and Connotation are the principle means (language) by which thematic ideas are translated into architectural reality. The basic how of this is covered elsewhere. Roughly, denotation is the what; connotation expresses associations related to colors, textures, forms and masses; together, they convey meaning. A building (or complex) that does not focus and express meaning is devoid of art and defaults to utility (at best), a mere building - it has no memory. This is not the purpose of architecture.



The Park is a place where someone could live and work for an indefinite period of time. The layout is designed to accommodate individuals, teams and groups in various combinations and with varying degrees of separation and privacy as required by their work. This “separation” function, however, must not translate into isolation nor does it prevent mixing in the Commons areas. The energy generated by different groups as they work must be, as is appropriate, be available to all. This is a scaling up of a principle long understood and facilitated in Taylor environments.

All the program elements required of a first class resort have to exist at Xanadu: sleeping, food, recreation, entertainment, support services and so on. These services and facilities will, of course, be positioned and delivered differently (than a resort) given the main mission of the Park and the requirement of seamless integration between creation and re-creation. Xanadu is a large structure but designed to be surprisingly intimate in scale. The square foot per person ratio is low. Spaces are design to provide a great “sense-of-space” while being, in fact, intimate in physical scope. For these reasons, and for reasons related to the program itself, the degree of integration of the “living” accommodations to recreation and work is very high. These are not isolated functions or places. Individual rooms, for example, will be set in suites of various sizes to accommodate teams when necessary. There is, given the program, a natural hierarchy in the relationship between these functions. This promotes interface and navigation and make up a major aspect of the environment “memory.”

The work areas are organized into four meta-process Zones: “Future World” - “Innovation Zone” - “Treasure Island” - “Voyage to Reality.” These Zones are large - each housed in a 150 foot diameter Dome that contains several levels and horizontal areas. While functionally dedicated, each of these Domes (Zones) can be configured to serve more than one group at at time, as well as, provide open “commons” dedicated to its specific function. These Zones will be oriented to the site in a way that reinforces their purpose: light, views, access to and from the Commons, and so on. Because of this orientation and certain site conditions (such as slope), the actual square footage of each Dome area varies and can be matched to its functional requirements. The intersections between each of the four perimeter domes and the center dome, create transition zones that can be configured in a variety of ways. One use is to expand the area of any one Zone as activity levels might demand. Another is to make full us of this transition element for facilitating migration between the various process Zones. Another, to set up support tools and services as might be apporopriate from time to time. The basic Zone-based architecture schema, as aoutlined, will form the floor plan of every OIP facility no matter its size or dedicated function (see: “There to Here”).

There will be total techological “transparency” between all of the areas of the Park and the outside world. This transparency can be “gated” as is appropriate. It can be configured right down to the individual level by each user.




The objective of this exercise is to create a clear “End State” vision that “drives” every intermediate design decision. Then, to define a series of intermediate projects that will progressively provide the OIP with greater and greater functionality and experience of “park-ness” until a full scale version is possible. Every element of the “End State” version should find itself, at some scale, in some way in every built iteration. This way the full “venue” of the Park is created in the beginning and develops, organically and by feedback as we progress.

Five projects are the minimum required steps to accomplish the “End State:” Inside KnOwhere , RDS Inside, RDS Snowflake, Park One and Xanadu. This is not to say that only these five will be built. Many versions of any of these steps may be executed. However, it is the five, each an order of magnitude more complex than before, that constitues the innovation path to the full concept.



Inside KnOwhere

8,000 sq ft

Inside a
20,000 sq ft



Environment for Park simulations.

Population: 75 - 12 hour use; 4 - 24 hour use.



RDS Inside

6,000 sq ft

Inside a
loft space



Office and work process area for Star Team and ValueWeb members. Park Simulations.

Population: 50 - 12 hour use; 4 - 24 hour use

Business Team functionality. Built inside an existing structure.



RDS Snowflake



40,000 sq ft

Stand alone

two acre

close proximity
support facilities

There is no reason that a Snowflake strategy cannot be scaled up to a “Park One” level.



Deployable environment with full simulation capability - can be developed in three stages. All functions can be represented and supported to a limited degree.

Population: 200 - 12 hour use; 50 - 24 hour use.

Profitable operations and feedback on venue. Facility can be moved allowing temporary use of land.

Test site for Xanadu energy systems.

England or France. May use existing Country Manor.


Park One




100,000 sq ft

Can be
stand alone
be part of an
existing complex

Requires access
to Urban Setting




Full functionality, limited scope of some venue items.

Population: 400 - 12 hour use; 100 - 24 hour use.

Park metaphor a reality, test of critical mass thresholds - permanent facility.

Location unknown - can be part of a larger complex.



600,000 sq ft

Stand alone
self suficient

Country setting



Full venue and expression of the concept.

Population: 2,000 - 24 hour use.

Total innovation environment - evolving facility. Stand alone on minimum of 20 acres.

Assumes countryside two hours (optimum) from Heathrow Airport.



These numbers are order of magnitude estimates and not fixed. The “RDS Inside” could be a little smaller or twice as large, for an example. However, the basic size and costs outlined are necessary for the functionality that is described.

An alternative to an RDS Inside, is to secure a traditional building and bridge to RDS Snowflake clusters as illustrated on the left. This strategy has many interesting ramifications and can work where the opportunity is to hold an historical house or church structure and develop the property over a period of time.

The high Capital costs associated with Park One and Xanadu can be underwritten by ValueWeb membership developed during the RDS Snowflake phase. Throughout all phases of development, it is important that the IEP Team see themselves as Park “designers” and “keepers” not, exclusively, “makers.” The ValueWeb members will be the primary makers and users. It is anticipated that many materials, products, services and technologies will be donated by ValueWeb members for R&D, demonstration and feedback, and promotional purposes. This will reduce the capital costs associated with the various projects.

This architectural program, therefore, develops in concert with feedback to and from the business program. It is, essentially, a “bootstrap” process. Build the tool that makes the value that enables a better tool that creates more value that...


Specific Architectural Elements That Can Migrate
Through Each Phase:



Meta-Process Layout with Center Dome:
For “KnOwhere Inside” simulations the existing KnOwhere spaces can be set up to demonstrate the “Know HOW Place” functionality. This includes the incubation deck. The Simulation to be run the week of July 9, 2000, in Palo Alto, will be the first application of this schema to work a real business opportunity.

Live-in Facility:
Palo Alto KnOwhere already has shower accommodations and an modest kitchen - Hilton Head will have these same facilities when upgraded. This means having some live-in capability is possible on a test level - Most likely for core-team and network knowledge-workers. The AI CubeOffice System can be employed to construct these personal areas and the will add, materially, to the functionality of the Stores. Early experience with how the live/work interface works is better than later - that are also pragmatic and financial considerations given the need of the team members to do extended travel.

Prototyping Capability:
“Inside KnOwhere” can be configured to support conceptual, programming, bread-boarding and model making levels of prototyping that can be done on a real-time basis.

RDS Inside

Meta-Process Layout with Center Dome:
This environment is the most constrained of the set - However, the basic schema can still be reflected in the layout and reinforced with strong graphics. A loft area can be employed if the ceiling height allows. As the principle use of the “RDS Inside” environment will be to house the Star Team and their ValueWeb members for continuing the design Development of the IEP, there is a certain self-reflection provided by this arrangement. Using AI Radiant Walls and dome elements (Cambridge and CHP), some sense of the “End State” shape can be accomplished.

Live-in Facility:
This will be about the same with the “RDS Inside” possibly more constrained by square footage and lack of other amenities in the space. Nevertheless, a good element to keep, if possible.”

Prototyping Capability:
Should be the same as the “Inside KnOwhere” facilities - may be a bit more constrained by square footage.


Meta-Process Layout with Center Dome:
This will be the first literal expression as the Xanadu layout can be accomplished with this building method and on this scale. This will be the first real test of both the schema, and the architectural expression of it using domes. In terms of scale, the RDS Snowflake Domes will be about one third the diameter of those of Xanadu. In terms of volume, this is a one ninth expression - about one order of magnitude smaller. It will be, however, a robust and effective expression of the idea and its functionality.

Live-in Facility:
The Pods make a good framework for living, bathing and food service facilities. These would be small, simple spaces as envisioned in the original 1986 Knowledge Factory Concept. These can be quite retreat and sleeping areas, not the full, self-contained environments envisioned at Xanadu. However, the RDS Snowflake iteration affords a useful application of this feature. While the use of full towers is beyond the scope of the structure, mini towers with three to four units on them are feasible along with “EcoPhere”-type applications. This can add to the total experience, bring more variety and playfulness into the design concept and provide useful experience for future designs.

Prototyping Capability:
Can be extended to serious modeling with a medium level metal, plastics and woodworking capabilities. Full print facilities and box making for package design.

Park One

Meta-Process Layout with Center Dome:
With “park One,” the scale will be greater than the “RDS Snowflake,” however, the shapes will be more dependent on the structure it finds itself in. If Park One is housed in a conventional structure, the idea will be to use that structure as a “box” with, abundance of side and top light source, and build a completely different environment within it. This will work, if the “difference” is translated to the vertical dimension, as well as, the horizontal. Vertical layout is under explored in architecture leaving the field to a “flat earth” set of solutions. To fully make a “sense of place,” many intimate factors have to come together. Not too hard to do in the home - in the workplace, however, intimacy and variety are usually designed-out of the commercial equation leaving little to make place. In a multilevel box, a rich interior can be crafted of light weight materials that can be manipulated like a stage-set, thus, overcoming the inherent limitations of the box. In this kind of situation, circular forms and domes can be introduced.

Live-in Facility:
On option with “Park One” is to build as part of a larger Urban complex including a hotel and other recreational facilities. In this case, live-in facilitates can be part of this larger complex. However, access, the exact nature of the rooms layout and other design aspect are key. Again, this is an interface issue. With the Cambridge KnOwhere Store facility we were next door to a very nice Hyatt Hotel. However, even this brief separation “broke” the continuity of the experience and put users back in a too conventional world. We were considering, if we had kept the facility, connecting the KnOwhere Store to the Hyatt with a glass enclosed bridge on the second level directly into a suite of rooms that we furnished with our own furniture. As we were the Hyatt’s single largest customer, they were seriously considering this option. An arrangement like this may be the best tradeoff for “Park One.”

Prototyping Capability:
About the same as “RDS Snowflake” - the setting make make this more difficult in terms of codes, access and costs.


Meta-Process Layout with Center Dome:
With Xanadu, the layout expressing and supporting the work process will find a fuller expression than the earlier projects. This will not be limitied to the basic meta-process Zones, but will extend to every process within the park. Indeed, the experience of Xanadu will be like traveling through a living mind. This is an interface design issue.

Live-in Facility:
My Bay Area Studio design is the most similar architectural concept to these facilities. The Studio project is well documented and should be seen as a precursor to this aspect of Xanadu. The complete and full integration of living facilities with the work environments is a key aspect of the Know How Place.

Prototyping Capability:
Full service/product, one-off, prototyping capability. Everything needed to manifest idea to form - the dichotomy between design and implementation broken.



Xanadu has a myriad of processes - formal and informal, directed and spontaneous. These processes are chunked into meta-processes - many of which are organized by physical location. Each of these process have a specific architectural expression, places where they take place, and configured tool-sets to support them.

These processes fit into an over all process Model that spans idea creation to the making of Intellectual Capital. For an example of this model-type see: Subsystem 6 of “System and Method For Augmenting Knowledge Commerce.” Many forms and sources of IP, from many organization and individuals, will eventually make up the formal processes of the IEP. A feature of any specific Park will be the close-coupling of the IP, the processes, the physical and virtual environments and the tool augmentation systems that support it all. This close-coupling, itself, is IP. All OIP environments will be “engines of creation.” Xanadu, as described, is one expression - one design - in a series of designed and managed (Design?Build/Use/Feedback) artifacts that will be created to augment and accelerate the innovation process. The innovation process, itself, will be used to create Xanadu.

No such integration, of such a wide range of processes, has ever been attempted as a formal system and on the scale of Xanadu. It is the goal of Xanadu, as a specific artifact, that this be accomplished with near total transparency. Xanadu will not “feel like” a managed environment on the level-of-experience of its individual users - this is an important principle for facilitating emergent “Group Genius.”. Much of the success of this seamless integration rests on the quality of design brought to the processes themselves, but even more so, to how well the physical environment and support tools explicitly support each process while maintaining transparency (Brin). This is an “Interface Design” Issue.

Processes are more fluid and mutable than technology which is more adaptable than architecture. To retain close-coupling over time requires a design strategy that takes into account the different mutation rates of these three major systems (processes, environments and tools). There are several aspects to this design strategy: de-coupling systems while integrating them; appropriate clustering of functions in non-competing levels of recursion; building an artifact of dynamic memory; organizing all components by amortization cycles are among them.

An example of de-coupling is making use of extensive “plug and play” between all components and never burying inside the structure anything that will have to be altered in the future. Functions have to be “closely coupled,” systems have to be integrated, units and components should be separated for easy recombination and exchange-out. A simple instance of this approach is a typical video conference room as distict to MagicWindows which brings the capasity to you.

An example of appropriate clustering of functions in non-competing recursion levels is...

An example of building the (total) artifact as dynamic memory is making every part, sub-component, component, subsystem and system that makes up Xanadu an AGENT. As an agent, it “knows” what it is, where it is, what it does, and what it’s prior experience has been. It is a rule-based system that “learns” from its experience and adjusts it’s performance through feedback. These Agents communicate with one another (by rules) and negotiate. In total, the are the memory of the environment. This method takes Brand’s philosophical perception of “How Buildings Learn” and adapt, and makes it useful in practical terms. This is accomplished by creating a language which codifies the experience. It is also possible, using this agent architecture, to build economy into the actual structure itself: each agent can represent the real artifact, know it’s state and status, and transact accordingly. Xanadu is an economy and an ecology.

A simple instance of this approach is for Xanadu to learn what specific combinations of temperature, humidity and light an individual finds comfortable (in a matrix of different activities and circumstances) and to adjust these factors (as possible to the mechanical limits of the system) for each individual as they move about the space.

An example of organizing components by amortization cycles is provided in the Program Statement of the Bay Area Studio Project.



Matt Taylor
Palo Alto
June 4, 2000

SolutionBox voice of this document:

Team participating in this iteration of the project:

MG Taylor Corporation Core Staff

AI Core Staff

KnOwhere Stores Core Staff

Yolke Core team

HP Star Team


Matt Taylor 843 671 4755


posted June 4, 2000

revised July 11, 2000
• 20000605.175659.mt • 20000628.131055.mt • 20000621.35615.mt • 20000628.124227.mt •
• 20000701.201525.mt • 20000711.61424.mt •

note: this document is about 75% finished

Copyright© Matt Taylor, HP Star Team MG Taylor, KnOwhere, 2000

IP Statement and Policy

Search For:
Match:  Any word All words Exact phrase
Sound-alike matching
From: ,
To: ,
Show:   results   summaries
Sort by: