Ideaspace

Ideaspace is a web-based service in which users can quickly create, modify and share ideas. An individual can create a flexible web-based space to hold textual and visual ideas. These spaces can be shared with other users who can leave suggestions and comments on the ideas. The space also contains creativity tools that help the user elaborate on the ideas they’ve created.

No longer are ideas strewn about on post-it notes or forgotten. For those who rely upon that chaos (which some creative types certainly do) this tool would certainly change the process of idea generation. Like other cognitive artifacts, Ideaspace would alter this process but not blindly. In reviewing research on creativity, we attempt to identify the rough tasks involved in idea generation and build a tool that supports those tasks.

A person who owns or works on one or more computing devices (PC or PDA), has internet access and uses these devices to store personal information might be a typical user of Ideaspace. Currently these users commit their personal information to proprietary formats that remain on these local devices. Ideaspace would allow a user to easily create a web space where this information could be uploaded and stored in an open format. The user could keep the information on the server or download it back to one of these devices.

Creativity

Creativity is often considered a mysterious topic and used to describe people who work in similar endeavors – artists, designers and writers. However, research from various fields recasts creativity as a process. Creativity is a series of activities rooted in an individual’s interaction with artifacts, the environment and other individuals. This view of creativity opens up the field as something not tied to one type of individual, person or profession – something that anyone can do.

Cognitive science research on the subject focuses on modeling creativity. The research is less concerned the environment and social interaction. Ronald Finke’s Geneplore model, one I take as an overarching framework that can include other lines of research, is composed of two stages – generation of ideas and later exploration of those ideas.

Idea generation begins with "preinventive structures". They may be novel visual patterns or verbal combinations. They may be ambiguous or vague at the time they are created. This may indicate that the structure has an implicit meaning to its creator. It might also happen than meaning that ‘emerges’ through the later process of exploration. Finke argues that the ambiguous and emergent properties of these structures are a hallmark of creativity.

Finke identifies various generative processes that might result in these structures. The most basic are the retrieval of existing structures from memory and the formation of associations between these structures. Another involves the synthesis of concepts. Single concepts can be combined to form more complex concepts, with the meanings of one or both of the initial concepts being altered as a result. Analogical transfer where a relationship or sets of relationships in one context are transferred to another can result in these structures. Categorical reduction can result in these simpler ambiguous forms - where a familiar object is conceived in only the most basic functional or geometric forms.

Exploratory processes help further the preinventive structure. Many of these processes involve looking at the structure from different perspectives. Context shifting involves considering the structure in a new light as a way of gaining insights about other uses or meanings of the structure. Functional inference involves exploring the functional application of an inventive structure. This begins by "imaging oneself actually trying to use the object in various ways."

Creative ideas seem to involve associative, analogical or random variations at either the generative or elaboration stage. Philip Johnson-Laird shares Finke’s view. "In short, creativity depends on arbitrary choices and thus on a mental device for producing, albeit imperfectly, nondeterminism."

Another interesting observation that Finke makes is the importance of creating intimacy. He sees intimate engagement as an important prerequisite for successful problem solving. One needs to be personally vested in a creative idea "in order to sense the full richness of its implications and possibilities." The private development of ideas surfaces in other research on creativity – specifically on group brainstorming supported by electronic tools.

Group brainstorming is often thought of as an efficient means to generate ideas because of the simple increase in people working on the task. Obstacles called production blocking arise because of the social dynamics of the situation.

Face to face meetings tend to create and consider ideas serially. One person typically holds the floor causing others to hold their ideas until the floor is open. This actually slows the number of ideas that can be generated in a session. The use of a shared electronic medium where all users can access and edit the same document does allow groups to consider topics in parallel. Production blocking is further reduced when participants compose several ideas in private prior to meetings.

Creative Process

From the review of electronic brainstorming and idea creation, different stages that form a process of growing ideas can be identified. Growing means taking an idea from an initial thought to a more fleshed out concept. This process eventually reaches a terminal stage.

There can be private and public parts of these stages. Private stages involve independent generation, creation, elaboration, and evaluation of ideas. Public stages involve the sharing of ideas to peers for elaboration and evaluation.

Private Phases
Generation – This stage involves coming up with the concepts that form an idea.

Creation - This involves initially composing the idea – in either textual or pictorial format.

Elaboration – This involves reflecting and attempting to look at the idea from different perspectives. During this process one should be able to edit and add to current ideas.

Evaluation: Looking over your ideas, making connections between ideas, perusing the idea space.

Group Phases
Elaboration: Once an idea is made public other users may comment on or add to the idea. This may involve a generative phase of new ideas that branch from the current one.

Evaluation: Determining whether the idea should be acted on in some manner. This phase is analogous to the private stage when a user is determining whether to make an idea public.

Alternate Tools that Support Creativity

Generating ideas and bringing them to fruition can involve private and public stages of composition, reflection, comparison, evaluation and elaboration. Since there are so many activities that take place, we must cast a wide net when looking at current tools that support these activities – considering tools from pen and paper to electronic mail.

Paper
The most familiar systems are pen and paper based – these can be post-it notes tacked to a corkboard or a personal diary. Regardless of the organizational method, these systems allow for the quick entry of ideas in either pictorial or textual format. A person has an idea and they jot it down. This is important to the earlier tasks of capturing ideas – when an idea hits you need to write or sketch it quickly. The interface of pen and paper is nearly transparent for this task – one can concentrate on the task of composing the idea without to much thought to the interface.

Paper based systems are also excellent for the private stages of idea composition. Once an idea is jotted down, it’s very easy to elaborate upon by revisiting the same piece of paper. The act of writing or sketching in a notebook is a very intimate experience underscoring how well the environment supports these private activities.

However, the medium can suffer from organizational problems. Once written down, those slips of paper must go somewhere. Some stricter organizational strategies may be to put the idea on a corkboard/whiteboard, in a notebook, or in a folder. Looser organizational strategies may be to put the piece of paper in a pocket, on the front of a computer monitor or in a stack of papers. The choice of a good strategy is crucial to how well the tool might support the remaining tasks we’ve outlined – reflection, comparison, evaluation and elaboration.

Organizing paper artifacts on a wall supports the later tasks fairly well. Essentially a corkboard is a large two-dimensional display. One can move away from the board to see the big picture – reflecting on groups of concepts. One can move closer to the board to see the small picture – reflecting or evaluating a single concept. If one wants to compare two ideas, one can reposition the ideas next to each other or simply scan between the two. If one wants to elaborate on an idea, one simply adds more information to the by attaching another piece of paper.

Diaries and notebooks are organized in a codex style – one page after another. To elaborate on an idea, one must write in the margins in the notebook or compose in a different location in the notebook. While not as elegant as a spatial organizational system, notebooks are portable and more supportive of the earlier phases of idea generation.

To compare or reflect on ideas one must flip through a notebook. The best-organized notebook is a chronological organization of ideas. To access old ideas one has to rely upon memory or this organization scheme. A more dynamic scheme of organization might better support the task of comparison and elaboration.

Both paper systems support social evaluation and elaboration. The corkboard scheme creates a space for discussion. The contents of a notebook can be shown to others, but often the contents are transported to a more transient representation – like a whiteboard. Neither system captures a discussion – we rely on a transcriber to capture what was said.

Email
An electronic version of jotting ideas down on paper is email. In the right kind of environment email approaches the same ubiquity as pen and paper. To a student on a modern college campus or a knowledge worker a networked computer is as close as the nearest lab or desk.

Email is often considered a communication tool, but like many technologies it can be used for different purposes. Emailing oneself creates an environment that supports many of the private tasks of idea generation and elaboration. If one has an idea and needs to jot it down, email provides a way to quickly compose text and save it. The interface – while not as transparent as pen and paper – is familiar to our target audience. Also, most knowledge workers will run an email application once while they are at a computer – if not for the entire time. The tool is readily at hand.

Why would someone email an idea to himself or herself? If configured correctly email is something that can remain on a centralized server and be accessed from anywhere with Internet access. This creates a somewhat portable tool like a notebook. However, once messages are downloaded to a local client they remain there - unless the user takes the time to resend each message to themselves.

Like pen and paper, once an idea is created in email its important that it goes somewhere – into an organizational strategy. Email brings it’s own strategy – the list view. This strategy is very flat and modal. If you want to compare ideas, you can open two messages and compare their contents side by side. Beyond that, you have to rely upon the relation of concepts through where they appear in the list. Messages are automatically dated and author stamped. Some of the more advanced organizational features of email depend on the email being downloaded to a local client. When downloaded to a local machine, a user can associate ideas by placing them in the same folder. Depending on the email client, messages can also be searched by keyword. But both of these features limit the main benefit of email as an electronic tool – its ubiquity.

For the social stages of evaluation and elaboration, a user has to email another person – or persons the idea they’ve been working on. If both users are online and have an open email client, the discussion might be near real-time.

Unlike notebooks and corkboards, a record of the discussion is automatically generated. Thus the elaboration and evaluation of an idea takes the form of a message being sent and the replies that follow. If a discussion is in depth, the resulting message can be long and possibly off topic. The final form might resemble an inverted pyramid – with a single idea at the bottom of the message growing into a wider discussion. Ideas can be spread out over many messages, email accounts, and clients.

Social Collaboration Tools
Threaded discussion groups are distant cousins of email. Both tools share a similar textual format that could provide for the capture of an idea. Both systems share a similar organizational structure – messages that contain ideas are time and author stamped. However, since discussion groups are hosted on a central server, users typically don’t have to worry about losing their ideas. Messages are also typically asynchronous.

Unlike email, a discussion group is inherently social. As a tool, this fails to support the needed privacy involved in idea generation. These groups may be very large and people might be concerned about publishing their ideas to such a large group of strangers. Discussion groups tend to suffer from some of the production blocking inherent in group brainstorming. For large groups people might spend more time reading messages than elaborating on their own ideas or generating new ones.

Oblique Strategies
The tools already discussed all address the later stages of the creative process. This is due to the difficulty in supporting a task that relies heavily upon random, associative logic. Two tools that do this well are, strangely enough, decks of cards.

Both Oblique Strategies and Creative Whack Pack are used like any other deck of cards – they are shuffled and then dealt. The cards suggest ways to approach dilemmas or creative blocks in a way similar to the generative processes described by Finke. The cards are dealt to oneself, much like an intimate game of solitaire, and the suggestions are considered.

Brian Eno and Peter Schmidt, authors of Oblique Strategies, say:

These cards evolved from our separate observations of the principles
underlying what we were doing. Sometimes they were recognized in retro-
spect (intellect catching up with intuition), sometimes they were formulated.
They can be used as a pack (a set of possibilities being continuously
reviewed in the mind) or by drawing a single card from the shuffled pack
when a dilemma occurs in a working situation. In this case the card is
trusted even if its appropriateness is quite unclear. They are not final,
as new ideas will present themselves, and others will become self-evident.

Some strategies may be nonsensical one word statements like – "Ghost echoes" or "Idiot glee". Most are longer attempts to jar the individual to rethink an idea from a different perspective – "Are there sections? Consider transitions", "Work at a different speed" or "Turn it upside down". If we try to categorize the strategies, they tend to fall into those dealing with the visual or aural. That makes sense –a musician and painters are the authors.

The Creative Whack Pack takes a more categorical approach by dividing the deck into four suits – the explorer, the artist, the judge and the warrior. Don’t be put off by the cheesy names. Roger van Oech is really trying to match a personality to what he thinks are parts of the creative process - exploration, transformation, evaluation and implementation. A person may be more willing to "Think like an artist" than try a "Transformation Strategy". Example artistic strategies might be to "Simplify", "Rearrange", or "Imagine you’re the idea". An exploration strategy might suggest you to "See! Hear! Taste! Feel! Smell!", "Look to nature" or "Change viewpoints."

For the private generative phases of idea creation, these tools try to make explicit the random analogical processes Finke described.

Design Rationale

Given that there are a number of good pre-existing tools that support creativity, what can Ideaspace do that is an improvement? What does the technology buy us?

One thing brought out in our discussion is that no tool tried to address the development of ideas as a long process. Oblique strategies were good at an early stage. Collaboration tools at a later stage. Ideaspace can address the entire process by offering a suite of tools loosely structured around the tasks I’ve proposed.

Environment
Our feature list of Ideaspace begins with a discussion of the environment. The environment is essentially a two dimensional plane that can be zoomed-in and out on. This is essentially based on the corkboard /whiteboard discussed earlier. This allows for the flexible viewing of ideas and the spatial arrangement of ideas. Tools to navigate through the space, perform actions upon a selected item are located in drawers that open when clicked. This maximizes the real estate for the display of ideas.

If one wants to view the entire space, looking at the whole and how ideas might be organized, one zooms out. At this viewpoint, ideas are represented at the lowest level of detail possible – a headline or short summary of contents. The relative size of an idea and its color denotes the size of its contents and the idea’s activity. If one wants to view an idea in more detail, one zooms in. The closer one gets to a group, the more detail that is presented about an idea.

The open structure of the environment allows the user to quickly type in ideas, move them next to other ideas or group them around explicit headings. This allows for spatial relationships between ideas to be built – important for reflection and comparison.

Ideas
An idea is limited in length (we're not talking about a 10 page paper - more like a 3 paragraphs), but unlimited in the way it can thread to other ideas or a discussion. From the discussion of paper and email, the recording of ideas should be fairly unstructured. This allows for ideas to be quickly recorded. Structure and organization comes as the idea is elaborated. Naming or categorizing ideas should be optional. A user should be able to be postponed indefinitely.

The interface for creating an idea should be flexible. In the same way you might jot a note on a scrap piece of paper, you should be able to send an email message to your Ideaspace for idea entry, for example. This allows users to have multiple ways to record ideas that are stored in centrally organized space.

If a user chooses, they can denote associations or relations between ideas. Ideas become linked when they are moved next to each other – this relationship exists as long as the spatial arrangement lasts. Searching can also form relationships. The relationships formed by searching allow for inferences and odd connections. Should this be at the level of the idea object - or should it be higher. Maybe you just have a bunch of scanners that know what to look at in other spaces for you.The manual linking could serve two purposes - to force a connection between ideas that aren't. This could create creative 'friction'. Also as a way to thread a discussion around an object.

A Sticky Space
The Ideaspace must maintain contact with its users at an interval agreed upon by its users. As we saw with other communication tools, the amount of contact a person has with an information space can vary from the nearly synchronous, as with email, to the asynchronous, as with discussion groups. Users are notified at 6, 12 or 24-hour intervals via email of what’s gone on in their space. This allows for asynchronous communication to go on in the space without the user constantly having to check for updates. Also, if the users are in a space at the same time, they can communicate synchronously with a chat applet.

Creativity Tools & Oblique Strategies
The creativity tools allow the user to elaborate a created idea. The creativity tools are essentially strategies that force the user to take a different perspective on the idea they’ve created.

One form of these tools can be the integration of a search engine throughout the interface. This might happen at a global level or a local level of the idea.

Idletime Creativity
At the global level is the environment of the space. Aside from simply being a static container, the environment can also act as an active tool to support idea generation. Starting out the environment is empty. If the space sits idle without interaction from the user, the user might be prompted into enter in headings they find relevant. These can be simple terms like "computer", "color" or "tempo". From these keywords, the tool performs a search – either through a local database or Internet search engines. The results are presented on screen in clusters around the initial idea. If the user clicks on one of these results, another search is performed. Through this dialogue the user can work from the most general to more specific knowledge or associative knowledge.

Scanners
Another global level that a search engine can operate is between Ideaspace. Agents can search other spaces for similar ideas. This can create a seamless connection between ideas and users.

At a local level is the visual search. The idea space can contain visual elements as well as textual ones. If the user selects an image and then performs a visual search, the results are displayed in clusters around the image. Algorithms that look at components of the image perform searches. In a sense, the user gets to see other visual pieces that look slightly like theirs but different.

Reflectors
Reflectors are the visual way that creative prompts like Oblique Strategies are represented next to ideas. A user would select an idea and then select "Reflect". Randomness of strategy and presentation is important with these tools.

Collaboration

The social aspect of the tool allows other users to evaluate those ideas. We rely heavily on others we trust and respect to sound our ideas – they provide feedback and comments. When a user shares their Ideaspace, they allow others to look and comment on their ideas. The social evaluation helps generate other ideas and engage in evaluative processes that filter and refine ideas.

Discussion takes the form of a discussion thread that branches visually from an idea. Since threads are formed around an idea, the actions a user can perform are slightly different from other social collaboration tools. Aside from being able to link to files and other ideas, the user can "Comment", "Question" or "Build" on an idea. While not technically different from the "Reply" command in most email programs, this terminology might focus the user on evaluating and elaborating the initial idea.

You must be able to bookmark ideas from other spaces and have them appear in your space. This is an offshoot of the sharing concept – you could bookmark an idea you see in someone else's space and have it appear in your space. You'd be able to watch the idea grow - follow the ongoing discussions and contribute to it.

The social rules of Ideaspace are simple. In your private space, you have the power to upload, create and edit files. An Ideaspace is analogous to your own private whiteboard or notebook – you have the control. Once you share something – make it public, you have to abide by the etiquette of the group that you’re sharing ideas with. If an idea is made public and other members subscribe to it, comment on it, in a sense "Add to it" the idea can’t be deleted by you. It still remains your property, but you can’t delete it from other’s who are still looking at it.

Storyboards

Storyboards roughly sketch a user manipulating the environment and using a creative tool.

	The environment is essentially a two dimensional plane that can be zoomed-in and out on. This allows for the flexible viewing of ideas and the spatial arrangement of ideas. Here, ideas have a uniform size and are explicitly arranged around headings.
	"Hide Headings" does just that. A view that shows just the ideas.
	"Relative Views" displays a zoomed fisheye view of the ideas. Ideas grouped around heading 2 are shown with more detail. Also, ideas are displayed with relative sizes – denoting the size of their content.
	"View as Stream" shows a linear presentation of ideas arranged around a heading.

	Select Heading 2 and "View Collapsed" shows a uniform presentation of ideas around the heading.
	Select Heading 2 and Heading 3 and "View Collapsed". Select Idea 4 in Heading 1 and "View Threads" shows the discussion threads attached to an idea.
	Select Idea 4 and "Zoom". Shows more detail of content in a thread.
	"Show map" presents a scale map next to the zoomed area.

	Drawers contain tools. Hidden to maximize the display of ideas.
	"Map" drawer that contains actions to manipulate the views of the space.

	User moves mouse to ‘Tools’ panel. Selects ‘Oblique Strategies’ and drags to idea.
	A strategy ("Emphasize") is presented as a layer over the current idea – which happens to be an image. The strategy on the layer begins to rotate.
	User clicks ‘Zoom’ icon.

	User moves mouse to ‘Tools’ panel. Selects ‘Oblique Strategies’ and drags to idea.
	A strategy ("Binaries") is presented next to the current idea – which is text. The strategy begins to animate.
	"Binaries" asks the user to categorize the idea according to simple binary oppositions.