Bubble diagrams are a useful design and analysis tool.  They help us understand space adjacency, layouts, and relationships, traffic flows, and relative space sizes among other things.  They are a visual tool for representation.  We can use them to help us understand aspects of the design problem we are trying to solve without investing huge amounts of time with complex and detailed drawings.  They can also be useful for ideation.

As a general rule, bubble diagrams are created after space adjacency analysis.  There should be one bubble for each space listed on the adjacency analysis.  The lines connecting those bubbles or the lack of lines are the depiction of the results of the space adjacency analysis.  Dark, heavy lines represent close or high adjacency, dashed lines may represent some adjacency, and no line obviously represents a lack of adjacency.  Sometimes in may be useful to actually sketch out a bubble diagram to help think through the space adjacency analysis.  Seeing the spacing represented as bubbles may help you to think through the relationships between spaces and make decisions as to how strong or weak the adjacency should be.

There are other things you can do with a bubble diagram to make it more useful.  You can draw them on top of a copy of the base plan.  This will give you some preliminary ideas about how the spaces will actually fit into the area you have to work with.  As mentioned previously making the bubble sizes larger or smaller to show the amount of relative space each will require is useful.  You just have to remember that the actual space layout is not likely to be a circle.  I have mentioned another technique in a previous post [Thumbnail Bubble Diagrams – A More Complete Portrayal] where you place a graphic such as a picture or drawing inside each bubble to represent what is actually going to be inside that space.  This technique simply provides an additional layer of visual detail to help with your analysis of how the space will function together as a cohesive whole.  It can also be useful if you are going to show the bubble diagram to the client for clarification or feedback.

With all of the potential benefits of bubble diagrams it is important to keep a couple of things in mind.  First, there are a lot of “rules” dictating how bubble diagrams should be drawn.  For example, no bubbles should touch or overlap.  No line should cross another line or another bubble.  These constraints are intended to make sure your bubble diagram makes sense logically and that the spaces flow or connect in a consistent manner.  It does not mean that your design implementation will follow those rules.

Second, the bubble diagram is a design and analysis tool.  As such you should validate you input into it’s creation and the output from how you use it.  As stated above, the bubble diagram is usually based on the space adjacency analysis.  You should use the space adjacency analysis to check off or validate that you have address every space and every adjacency and non-adjacency.  The same applies going on to subsequent steps.  You might do an overlay of your bubble diagram over various form compositions.  That in itself is a validation process.  The next most likely step is to create functional diagrams.  Again you should make sure that your functional diagram carries forward the adjacencies and other relationships expressed in the bubble diagram.


A concept that I have been intending to write about for some time is “parti”.   A parti is usually a sketch, diagram, drawing, doodle, or some other graphic that represents the direction, concept, or theme of a design.  The concept of parti is common in architecture.  It is also used in other design disciplines.  It is seldom mentioned in conjunction with landscape design however.  That is part of the reason why I have not written about this concept until now.  The other reason is that a parti is a vague concept.

A parti diagram does not necessarily represent what the design will look like when it is done.  It is usually not a polished diagram.   It can be very rough; the proverbial back of a napkin sketch.  Parti has been defined as “the big idea”, “the central concept”, “the essence of the design”, “the design approach”, “the core element” and numerous other ways.  In almost every case a parti is described as conveying the meaning, form, direction, essence, scheme, approach, or some other aspect of a design.  If you are confused about what a parti actual is, I was too initially.

The first thing that was unclear is when in the design process a parti is actually created.  The answer is that you create a parti after you have some analysis completed.  You have to know where you have opportunities and where you have limitations.  You have to know the client’s requirements.  You should understand what functionality you need to provide.  You should have created at least some bubble diagrams and prepared an adjacency analysis.  In most cases a parti is going to come after some level of form composition analysis also.  You may create several form compositions that you evaluate as potential starting points for your design.  That being said, creating a parti comes after having a thorough understanding of the site, the client, and the functional and spatial aspects of your design.

The second confusing aspect of a parti was how it fit into the creative or generative portion of the design process.  A parti is described as a vision and/or an inspiration.  A parti is also shown as being a result or an output of one or more design concepts.  Creating the parti comes after developing conceptual designs.  Your source or inspiration for your conceptual designs may come from the site, the surrounding area, the client, the environment, or some other source.  Your client may have a love of camping that leads you to develop a concept based on nature.  The client residence may be of a Spanish style architecture that leads you do develop a Mediterranean theme concept.  There a numerous possibilities.

So what exactly does a parti do?  Why should you create one?  I think a parti is a communication tool.  It communicates the intent of your design concept.  In A Visual Dictionary of Architecture (1995), Frank Ching defines a parti as “the basic scheme or concept for an architectural design represented by a diagram.”  The parti should communicate something about the form as well as the concept.  Ideally, your parti will communicate the experience you intend to create.  It should depict something about the functional, sensory, and/or emotional aspects of your design concept.

I am not convinced a parti has to be a diagram or sketch.  A picture, an object, maybe even a simple storyboard may serve the purpose of a parti.  Which leads to the second question; why create a parti?

Anything that we can create that will make conveying our design intent to the client easier and more effective is a good thing.  We all live in a world of headlines.  We are flooded with information.  We scan e-mails for important subjects.  We skim newspapers for headlines.  The 30 second sound byte is the norm.  Imagine the power of a diagram or simple graphic that you can show the client and they will immediate see what you want to do.  Maybe your plan view does that.  Or maybe you created a perspective illustration that conveys everything the client needs to know.  You may not need a parti in every design.  However, if you can create one, it would certainly add value to your client presentation.

There is one very important difference in how and why a parti is used in architecture versus landscape design.  In architecture the designer is working in a third dimension in creating a building or structure.  That is not to say landscape design does not involve height or structural elements.  The mass of a structure just does not impose upon our designs the way it does in building architecture.  This is why I think our use or interpretation of a parti can be different.

As I said earlier, a small storyboard may be what you need to convey your parti.  Maybe there was an object or something that you saw that inspired your design concept.  A picture of that object may be your parti or a part of it.  Maybe one of your form compositions can be modified to express more fully the design concept.  Again, what we are looking for is a communication tool.  The format or media does not really matter.

One last point about the value of a parti.  I have read in several places that a parti should “anchor the design”.  In other words, when a design issue or question arises, you should be able to go back to the parti for answers.  In other posts I have mentioned the value of graphic tools such as a client profile, journey boards, inspiration boards, etc. to facilitate the design process.  A parti can serve the same purpose.  It communicates the intent of your design concept to your client.  Having your parti in front of you while you are designing will serve as a constant visual reminder of your design intent.

When creating bubble or functional diagrams it becomes important at some point to begin working in rough scale so you can understand proportions and space utilization.  It is very easy if you are using diagramming software to make bubble diagrams or functional diagrams proportional.

Using PowerPoint as an example, you draw a rectangle four by two to represent forty by twenty feet or six hundred square feet (a one inch equals ten feet scale).  When you create the bubble shapes you size them according to the required space.  You can make any shape you want.  You can pick the best shape to represent each area.  You just need to scale it.  In PowerPoint you would right click on the shape, select format shape, and set the size to represent the scale you need.  For example, a circle representing a table that needs sixty three square feet could by a circle with a diameter of nine feet or .9 inches.  If you want to use a square that is proportional you could make a square with sides of 7.95 feet or .795 inches.  A rectangle would be set at .7 inches by .9 inches to represent the size.

The point of this is to emphasize an issue about scaling in bubble or functional diagrams.  Do you use a proportionate size for the object itself or a size for the space the object requires when being used?  A table is a good example.  The table, with the chairs pushed in may be seven feet by nine feet.  However, when guests are at the table sitting in the chairs the space requirement is more likely nine feet by eleven feet at a minimum.  A grill is another good example.  The grill itself may have a footprint of ten square feet.  If you are using the grill during a party or dinner the space may easily double to allow for room to maneuver and to avoid the heat and smoke.

Given the fact that you are working from your adjacency matrix, the proximity or lack of proximity may well be important in how bubbles or functions are placed within the space.  One technique that I use to help with this issue is to create a shape and then duplicate it in a larger size.  The original shape represents the footprint of the space or object.  The second shape represents the required footprint when it is being used.  I then simply center these shapes on top on each other with the usage footprint on the bottom.  For example, the diagram below shows the space for a table and chairs.  The inner shape is the object footprint.  The outer shape is the usage footprint.

Bubble or Functional Shape Adjusted for Usage

Not every object may require extra space.  You may also be able to adjust how the usage footprint relates to the object footprint.  For example, the functional diagram below shows the lounging area has the same usage footprint as the space itself.  The grill foot print extends out to the front and to the left side since the back and right side of the grill are at the edge of the space.

Functional Diagram with Shapes Adjusted for Usage

Showing the buffer area that will be utilized when the space is being used allows you to make adjustments that are needed due to space adjacency requirements.  Also, this is extremely useful in planning traffic flow, overall space utilization, and space allocation.

There is some additional work in going to this level of detail but it is fairly minor.  As long as you establish your scale in the graphic program you are using and work out the areas required for each functional area most graphic programs make it easy to scale the functional objects.  The effort may help you uncover a potential problem long before you ever get to the design stage.

Bubble diagrams are a great tool for ideation, adjacency analysis, and space planning.  In a previous post I stated that I think functional diagrams are a more formal tool to explore how spaces will fit together and get a better sense of potential design layouts and patterns.  Bubble diagrams are the rough tool.  Functional diagrams are a more precise tool.

In making the transition from bubble diagrams and functional diagrams it is useful to have some idea of form composition.  Are you going with a rectangular, diagonal, arc and tangent, etc.?  Having an overlay of the base plan that is marked with the various lines of force that you want to evaluate can be useful for building your functional diagrams.

Consider this example.  The owners have a barren back yard.  They want 600 square feet of outdoor space that accommodates a table for dining, a conversation area with a fire pit, a lounging area, a grill, and a water feature.  The space is open off of a sliding glass door.  There are windows on each side of the sliding door.  Shown below are to base plans with lines of force marked for rectangular and diagonal form compositions.

Rectangular Lines of Force Form Composition

Diagonal Lines of Force Form Composition

In the rectangular form composition there are only lines running out from the back of the house at the door and window openings.  However these three openings suggest a pattern extending from the back of the residence.  The diagonal form composition has twice as many lines since they run in both directions.  My take on this pattern is that the two lines extending from the door opening have the most potential.  My decision is to go with the rectangular form composition because of the three major lines extending from the back of the house.

The bubble diagram I prepared reflects the results of the space adjacency analysis I performed.  The water feature should be visible from the table, conversation area, and lounging area.  The table should be accessible from the grill and the house.  The grill should not be overly close to the house, table, lounging area, or conversation area.

Bubble Diagram of required spaces and adjacency

At this point I am ready to see how the bubble diagram interfaces with my form composition.  I am using PowerPoint in this case, so I simply copy one on top of another:

Bubble Diagram fitted to Lines of Force

Everything seems to line up but this may not be the optimal placement.  That is really the point I am trying to make about the difference between bubble diagrams and functional diagrams.  I want to explore placement of spaces within the form composition to determine my final design layout, spacing, and placement.  The bubble diagram was a rough tool.  It helped me establish adjacency relationships. I need to go a step further and begin placing functions within the space.  The diagram below shows how the functional “bubbles” were generally placed.

Functional Diagram overlaid on Lines of Force

The lounging area is positioned to the right since that area offers the best sun exposure and it is less likely to interfere with the grill.  The table and conversation area were positioned to the right, away from the grill.  The table is closer to the door.  The lounging, conversation, and table areas all have visual contact with the water feature.

The next step is to create the actual design pattern for the space and to physically position the areas more precisely.  This is the final design.  The two side pieces build off the lines of force from the windows and also push the lounging area and conversation areas further away from the house.  The water feature makes a nice focal point and is centered on the lines of force from the door.

Final Design based on build up of form composition, bubble diagrams, and functional diagrams

Using overlays to check patterns and explore ideas is easy.  DynaSCAPE can be used to create the base plan which can be exported to PowerPoint, Photoshop, or a similar program.  Even scanned images can be used.  Use may have to remove the white background color since overlays tend to work better if they are transparent.

Using analytical tools and techniques is all part of the validation process.  The preliminary bubble diagrams were based on adjacency analysis of the required spaces.  These were overlaid on to base plans for form composition analysis.  Finally the selected form was overlaid with a functional diagram to identify the placement and relationship of the spaces.  These preceded the preliminary design and final design.  Going through these steps helped assure that the design was appropriate and met the client’s requirements.

An interesting post in the Designers on Design blog today titled “Plan B“.  The thrust of the post by Danilo P. Maffei, APLD, is that only one design should be created for the client; there is no need for a backup plan if you know the first or primary plan is your best work and it is the right design for the client.  His argument is that not only does it take more time; it also makes us less committed to the success of our primary plan.  There are several interesting follow-up comments to the post.  The post and comments are well worth reading.

I believe the best approach is to have one single final design unless the client specifically asks for multiple designs and is willing to pay for them.  In this case, each plan should meet the same criteria in terms of meeting the client’s requirements.  Serving up two completely different designs that meet the same requirements means a substantial amount of additional work in terms of validating that each design provides the same functionality and meets the client’s needs.  The only way this could vary is if the client asked for two or more plans that provided different functions or were based on different budgetary or time constraints.

If there is a need for experimentation or consideration of alternatives, that should come during pre-design.  Frequently during the ideation phase I work through iteratively.  Based on some usage scenarios I try to understand the adjacency requirements and prepare a few bubble / functional diagrams.  Then, I will shift gears and start looking at potential form compositions.  After generating some ideas I will go back to my bubble / functional diagrams and see how they work within the form composition ideas I have generated.  I may start looking more closely at traffic flow or other issues.  Two or three of these ideas may be worth pursuing in more detail and may be considered as potential starting points for preliminary designs.  Preparing more than one preliminary design is acceptable and may be worthwhile from the standpoint of validating the client’s requirements.

The point of design validation is to make sure that all of the alternatives, choices, and issues are resolved before the final design is completed.  Completely validating a design implies that the one and only final design meets the client’s needs.  It should match the client’s style and tastes.  It should include the hardscape and softscape elements that the client prefers or will be happy with.  If the design has been fully validated there should be no need for a Plan B.

My last two posts have dealt with prototypes.  Prototypes are representations of the design objective or some portion of the design objective.  That being said, almost everything we create to show the client as landscape designers qualifies as a prototype.

In my first post on this topic, I mentioned that I have been a big proponent of prototypes for many years.  However, this was in the context of information systems.  I did not see an immediate fit for prototypes within landscape design.  In thinking about prototyping from the standpoint of how they are used and their varying levels of fidelity, I think the prototyping approach is actually used within the landscape design field.  It just may not be used consciously.  Let’s look at some examples.

First, assume you get a call from a homeowner who wants a consultation on what is feasible in their backyard.  They have limited space.  They would really like a pool but they need a deck or patio for entertaining.  They just don’t see how it could work.  Assume you meet with the client, make some measurements, and evaluate the space.  Since this is a simple consultation, you grab your note pad and sketch out a rough diagram of the space.  You use bubble diagrams to show how the functional areas, the pool and the entertaining space, could fit into their yard.  That hand sketch is a prototype.  You are using it as a proof of concept.  It has zero functional fidelity and very low visual fidelity but if it convinces the client, it did the job.

In our second example, you are designing a front yard walkway.  The clients want to enhance the curb appeal of their home.  Their current entry way is obscured and offers visitors no clue of how to get to the front door.  You take measurements, get the budget, and get other information from the client.  You create a colorize plan view showing the new front yard bedding and walkway with all new hardscape and plantings.  That plan view is also a prototype.  It has high visual fidelity in that it is colorized and shows the flow of the walkway and how it is visible from the street.  It has some functional fidelity in that it shows how it functions by opening up the view from the street.  However, it is not high fidelity because the clients have to translate from a 2D drawing to how it will look when it is actually done.

In this last example, you have already sold a design.  You are creating an outdoor entertaining area with a large fireplace at the end of the patio.  During the design phase there was a lot of discussion about size and cost tradeoffs and the client made their decisions about how much space they were willing to pay for.  Construction has begun.  The patio area is dug out.  The concrete will be poured in a couple of days.  After that, the construction of the fireplace area will begin which includes side benches for seating.  The client calls in a panic and says they just do not see how it will fit.  They are very concerned about how the fireplace and seating will fit and that there may not be enough space.  Realizing the problem is one of visualization you go to a store and get several cardboard boxes.  You go to the client’s home to meet with them.  You grab the cardboard boxes and set them up, one on top of another, to represent the fireplace and side benches.  You move the client’s outdoor table into a position to represent how it might be set up after construction.  The client can then see a physical object where their fireplace and seating are going to be located and how a table and chairs fit into it.  This leads to some discussion and playing around with the space.  The client decides they really want more space.  They are willing to reduce some of the other yard areas to get it.  They can afford the change and ask you to have the contractor push the fireplace back two feet.  They realize it is their decision and are willing to pay the additional cost and suffer the delay.  Those cardboard boxes were a prototype.  It had very low visual fidelity but moderate functional fidelity from the standpoint that the represented the scale of the objects.

Each of these examples shows how a different representation was used as a prototype.  The first two, the quick sketch of a bubble diagram and the colorized plan view are very common in landscape design.  Using cardboard boxes to represent hardscape is less traditional.  In each case, the degree of visual and functional fidelity varied but was adequate to achieve the objective.

In the first case, you could have taken the client’s information and gone back to the office to create a plan view drawing to show how there was adequate space for the client to achieve their objectives.  There may be cases where you have to do this to convince the client.  Some client’s want more detail or want more precise drawings.  You have to judge the client and determine what will work for the client and your objective.

In the second example, there are other approaches that could have been taken.  You probably would make the plan view drawing anyway.  Colorizing it was an enhancement that may or may not have been necessary.  The plan view would show how the space was opened up and how the new walkway was more clearly defined.  Colorizing just adds to the visual fidelity in that it shows space relationships between hardscape and softscape more clearly.  It may also show the client how you are working in their preferred color scheme.  There may be cases where the client just cannot make the 2D to 3D translation and creating a 3D walkthrough may be necessary.  This is again something you have to determine for each client and project.

In the last example, objects were used to represent hardscape objects within the design.  You may be need to get creative when you to try to physically model something.  Sheets, plastic sheeting, or large sheets of brown wrapping paper can be used to represent walls or fences.  They can also be laid out to represent walkways.  A trash can may be used to represent a water feature.  Hose or cord can be laid out to represent bedding edges.  There are endless possibilities.  You just have to determine which design elements you need to represent and then what you can use to represent them.

All of these examples of prototyping also demonstrate how prototyping can be applied at different stages of a project.  In the first example, it was used as a proof of concept.  The second example showed how it was used to convey the design result.  The third example showed how prototyping can clarify issues and/or determine if changes are needed during construction.

None of the examples dealt with prototyping as a means of facilitating requirements gathering.  This is probably one of the most useful applications of prototyping.  Being able to show clients a prototype representation of some portion of their design may help the client clarify what they want and allow the designer to explore ideas that the client has not expressly stated.  What you prototype and how you prototype really depends on what you are trying to accomplish.  Prototyping can answer questions.

Bringing in stakes and cord to mark off different functional areas can help the client understand how their new space will be proportioned and how it will flow.  If there are questions about how confining a fence or wall will be, one approach might be to hang clear plastic sheeting in one area and a more solid material in another to give the client a sense of how different materials can provide varying degrees of transparency.  That same question might apply to the fence itself.  Placing some plant materials in front of a sheet to represent how the fence is obscured may convince the client that the fence will be a good functional element but it can be made visually attractive though the addition of plant materials.

There are many, many opportunities to use a prototyping approach in landscape design.  I realize I am using a very loose definition of prototyping.  In other design disciplines, prototyping is more of a way of approaching a problem.  It is not necessarily any one tool or set of tools.

In landscape design, prototyping can be both an approach and a tool.  I have frequently recommended that client contact and feedback be a priority throughout the project.  This provides opportunity to show the client small prototype components of the design.  Getting client concurrence early in the project and often, helps prevent costly redesigns.  As the project moves along, designers should keep prototyping in mind as a useful tool to help clarify issues or develop understanding of the design intent.  More importantly though, the prototyping approach can provide validation of both overall design concepts and specific design elements.

The last several posts have dealt with high-level view of my design process model.  The series covered each major phase and how they build on and overlap one another.  Subsequent posts discussed how the phases overlap with the two major project inputs, the client and the site.  This post deals with the details of how specific tasks and activities fit into the phases and how they support subsequent tasks in later phases of the design process.

A few points need to be clarified first.  In the diagrams below, I have laid out the representation of tasks and activities in a timeline format.  That does not necessarily imply there is a rigorous schedule that must be adhered to.  In an iterative design model, you cycle back and forth between and within phases.  It may make sense in one project to start a task early, and in the next project that same task may get postponed.  So interpret the timeline depiction loosely.  The starting points for tasks are determined based on what would typically make sense and based on the completion of any prerequisite tasks.  The length of the blocks is again based on what would typically make sense.  Some tasks may take more or less time depending on the project.  Additionally, the timeline blocks are grouped by phase for ease of identifying tasks in a particular phase.  They are not necessarily aligned horizontally with precursor tasks in earlier phases.

Second, some tasks overlap more than one phase.  The diagrams below have a legend that reflects the phase of tasks by color.  Tasks that span two phases are shown in the color of the phase they start in to the left and the color of the phase they end in to the right.

Lastly, this is not a comprehensive list of every activity or tools available.  It encompasses more detail than the high-level model but it is not an exhaustive checklist.  Very specific tasks such as soil analysis or surveys are rolled up into high-level task descriptions such as site analysis.

The first diagram depicts the Client Interaction tasks.  These tasks and activities are color-coded blue.

Client Interaction Tasks

Client Interaction tasks start at the inception of the project.  The task, Client interview / Follow-up / Meetings, spans the entire project to stress the need for regular interaction with the client.  Several tasks span into the Data Gathering phase so they are color-coded blue at the start end and violet at the end.  The remaining Data Gathering tasks are next.

Data Gathering Tasks

These tasks are color-coded violet.  They tasks also begin with the inception of the project.  Some may occur during the initial client meeting.  Others may follow days later.  The actual scheduling will vary by project.  The next set of activities and tasks is Analysis / Synthesis:

Analysis / Synthesis Tasks

These are the follow-on activities and tasks that you will complete to analyze the data you have gathered.  This is where dependencies begin to arise.  You have to have the data before you can analyze it.  The tasks in Analysis / Synthesis build on one another.  You have to do some before you can work on others.  What you find may lead you back to re-analyze what to found earlier.  This is the iterative process.  When Analysis / Synthesis is far enough along you can begin Creative Development:

Creative Development Tasks

A substantial number of Creative Development tasks lead to the creation of Client Deliverables.  You will also note that they start while Analysis / Synthesis is still underway.  You may begin exploring design concepts while you are doing analysis.

The key document you produce during Creative Development is the design program, which cannot be completed until Analysis / Synthesis is complete.  You have to analyze all of the data and synthesize it into a meaningful set of client and design requirements in order to preduce the design program.  Much of what is done in the Creative Development phase is reused, as you will see when we add the Client Deliverable tasks:

Client Deliverable Tasks

The Client Deliverables are what you present to the client when you are finished with the project.  You will typically walk through these with the client during the final design presentation.  Not every project will require every item listed.  Some projects may require other deliverables.  Whatever the items are, these are the result of everything you have done to this point.  They are the basis for the client proceeding with the project and beginning construction.

These diagrams fill in the gaps with specific tools and techniques to support the phases in the high-level model.  Treating these tools flexibly and remembering that it is an iterative process is key.  You will most likely not use every tool in every project.  You have to choose which to use based on the scope of the project and what you have to work with.  The objective is to build confidence that you have discovered all of the client’s requirements and that your process and results support the logical thinking that evolved into your design concept.  The target is to create Client Deliverables that meet the client’s needs and satisfy the requirements of the site.  Following a logical process and working iteratively, you will be able to meet your objective and have a satisfied client.