Computer-Aided Participatory Design
Yoram Reich[i], Robert F. Coyne[ii], Suresh L. Konda[iii], Sean N. Levy[iv], Ira A. Monarch[v], Eswaran Subrahmanian[vi], Arthur W. Westerberg[vii]
15 December 1992
Abstract. Participation in design is an old
concept, nevertheless, it is hardly used in practice. There are many issues
that impede extending the practice of participation. Improving the
effectiveness of participation may benefit from the development of support
tools. Some implications of participation to computer tool design are discussed
and a prototype computer tool called n-dim that addresses these
implications is described with a brief example of its use in a hypothetical
participation project.
Keywords: participatory design, design support
system, modeling, integrated design system, computer supported collaborative
design.
May 2005
Since the writing of this paper,
topics such as knowledge management, collaborative tools for supporting design,
tools for aiding customer participation in design, ontologies for representing
decision making, and many other related topics have emerged and proliferated. More
than 12 years after writing this paper, there are still no breakthroughs beyond
the ideas presented here. If you find such advances, please let me know.
Yoram Reich
The potential role of computational tools for supporting participation
Participation is hardly used in practice
tools and techniques have been developed over the years to support the kind of
participation activities that have been pursued in the past. New computer tools
may add a new dimension to the suite of available tools and extent the scope of
participation possible
Since there are some successful
participation projects carried out in computerless environments indicating that
computers are not a prerequisite for participation [1, 2, 3, 4,
5, 6,
7, 8, 9] the question of the need for computer tools for
supporting participation arises. Our answer consists of five parts. First,
especially in product design, computer tools such as 2 or 3-dimensional
simulations have been used already in participation activities [10, 11, 12] and can be further used instead of
expensive 3-dimensional scaled-model simulations [13,
11]. Second, a major part in participatory design
projects is the education of participants. The medium of education may range
from workshops to the provision of working manuals [14, 15, 5, 16, 17]. This educational activity can benefit
from the availability of computer tools that may enhance or replace some of the
traditional techniques. Certainly, hypertext and multi-media technologies can
be used to provide different, if not richer, medium for communicating
information [18, 19].
Third, extensive participation requires
facilities to enable the accumulation, organization, comprehension, and use of
the resulting volumes of information if it is to be digested before an action
is to take place. These facilities can be supported by computer tools. Note
that such facilities extend and provide new roles for other technologies that
facilitate group work such as Group Decision Support Systems (GDSS) [20, 21] or Group
Communication Support Systems (GCSS) [22].
Fourth, certain forms of computational
support may increase the rate and type of participation by providing effective
means of communication (see below). Fifth, and maybe most significant,
computational support for participation may provide the substrate for
conducting studies on participation. In fact, as the first step in a
participation project, prospective participants could study previous design
situations to determine the form of participation they want to adopt or modify.
We argue that the study of participation should be done in a participatory mode
(i.e., Participatory Action Research) and believe that the same tools that
support participatory design could be used for supporting participatory
research. In the ideal case, these processes are merged into one.
We now focus on computational support of
communication and information organization in participatory design.[viii] The participatory design process,
including the participatory development of tools, is a complex process whose
effectiveness may be enhanced by support from computational tools. Some
evidence that computer tools can influence participation can be adduced from
studying the patterns of interactions different people display via electronic
mail or other media systems [25]. For instance,
some people that did not participate in face-to-face discussions found it
easier to participate through electronic medium, even when participants
belonged to unequal social status [26].
Participatory design requires extended and
meaningful communication. Communication among members of the design group
requires that there be consensus on the naming (i.e., a shared denotation and
understanding of relevant terms and concepts), constraints, problem
structuring, and design trade-offs. Without such agreements, effective
communication and coordination of work cannot occur [27].
However, such agreements cannot, in general, be imposed from the outside but
must be generated by the design group consensually. In this context,
participants from different disciplinary, experiential, and organizational
background need to work together. Supporting the underlying communication
process requires facilities for reconciliation of different perspectives as
well as the maintenance of each individual perspective. Otherwise a given
individual participant may be unable to effectively access, understand, and
contribute to the information generated collectively.
The participation process takes place both
through synchronous as well as asynchronous communication. Communication, an
integral part of any participatory work, operates along the two dimensions of time
and place [28]. Interactions in the same-place
same-time quadrant of this space are the focal point of most collaborative
work. The focus of these tools is on creating a shared workspace where the
communications take place through a shared workspace and by the direct physical
presence of the participants. Extensions to these technologies, relax the
requirement of same-place by exploiting high-speed communications and video
technologies [29].
Communication becomes more complicated
when same-time same-place collaborative work, being resource and travel
intensive, is not viable and different-time different-place collaboration is
required. In the absence of face-to-face contacts, individual engineers must be
able to participate in this dialogue in an asynchronous manner in a variety of
representational forms, media, and modes of communication while retaining the
time sequence of the exchanges as follows:
·
information comes in a variety of representational forms including
sketch, picture, gesture, text (print and speech), graph, table;
·
information is exchanged in a number of media including paper and pencil,
physical being, computer, video, film; and
·
the representational forms exchanged in these media come in formal to
informal modes of communication: reports, memos, e-mail, face-to-face.
Information capture and structuring
depends on which representational forms, media, and modes of communication are
used. For example, information techniques useful for text management are not
useful for graphical information. On the other hand, organizational units or
concepts extracted from textual information can be useful in classifying
graphical information such as sketches, or drawings.
Same-time technologies use multi-form,
multi-media communication in an attempt to provide support for approximating
the most effective mode of communication - face-to-face. Same-time approaches
are limited, however, in the number of participants that can effectively
communicate. Furthermore, they lack a viable mechanism for accessing design
rationale or history, which causes significant problems in understanding
previous decisions and consequently in applying the lessons of the past to the
present. On the other hand, asynchronous communication is limited in the
richness of its modes of communication. However, it may support the
participation of larger numbers of participants and provide facilities for
capturing design rationale as an integral part of the process without
unacceptable additional overhead.
One of the major problems associated with
asynchronous communications is that the ``conversation'' tends to drift; i.e.,
the communicators tend to lose the overall context of the discussion [28]. This is because the maintenance of the structure of
the context becomes very difficult unless such maintenance is designed into the
system. For example, e-mail users attempt to maintain part of the context by
including the mail being responded to in their replies but a moments reflection
reveals the difficulty of using such an approach in an extended conversation.
Moreover, the ability of synchronous
communication to provide effective support for collaboration degrades when
meetings do not address self-contained topics. In situations where issues are
re-visited in subsequent meetings, the context of the previous discussions
needs to be re-created to allow effective communication. This activity
can benefit from facilities that assist in constructing and managing such
context -- which, in effect leads to problems and requirements parallel to the
asynchronous case. In effect, this brings different-time characteristics into
same-time activities.
Asynchronous communication is needed when
some cannot or are not willing to participate fully in all discussions, they
may follow the discussion coded through the computer medium and comment on
them. Their comments are a form of restricted participation providing input
otherwise inaccessible. Note that such an option is expected to be negotiated
and agreed upon by all participants. Asynchronous communication is also needed
when participation must be interrupted because of time commitments. One can
envision potential participants who are unable to participate due to conflicts
in schedules--a situation common in urban planning participation projects in
small towns [30]; in these cases, tools which are
designed to allow for asynchronous participation can permit those potential
participants to contribute. While the local newspapers have served as a record
of such participation [30], yesterday's newspapers
are sometimes hard to come by.
Existing computational tools and participation
Different computational tools are designed
to support the communication between group members to various degrees. Some
tools provide the basis for participants to communicate via a shared workspace
[29, 31].
Other tools support the communication process by recording the structure of
issues raised in the participation process (gIBIS [32]).
And still other tools can elicit issues from multiple participants, provide
feedback and guidance, as well as perform analysis to detect discrepancies
between different views (KSS0 [33]).
These tools and others do not have
facilities to support some of the participation requirements for tools outlined
above; such as, maintenance of contexts, threads, and perspectives.
Furthermore, all the above tools were developed for collaboration among group
members familiar with the use of computers, and independent of their
effectiveness in practical use, no assumption about the proficiency of users
with computers can be made with respect to participation in disciplines remote
from computers at best for the time being.
In our work on design research, the issues
and requirements for participation tools raised in this paper became apparent
in the development of support tools for engineering design. The issues of
design, implementation and incorporation of the design tools as an integral
part of the working of a design organization made us confront the issue of
participation and its scope. This research has been applied to the n-dim[ix] project [34, 35, 36]. While the
primary aim of this project is to build a computational support tool for
engineering design, this tool had to have similar properties to participatory
tools; these include maintenance and reconciliation of multi-perspectives,
autonomy of individual perspective and, maintenance of contexts and threads of
discussions and negotiations. These conclusions were reached on the basis of
empirical studies of engineering design conducted by some of us and by others [37, 38, 39, 40, 41, 42, 43, 44].
Some of the significant findings were:
·
Designers typically spend as most 15% of their time doing standard
analytical tasks, the rest being spent negotiating various aspects of
the design.
·
Communication between designers occur in a variety of forms (e.g., text,
gestures), media (e.g., drawings, video), and modes (e.g., formal meetings,
corridor chat).
·
Many errors in and failures of design occur due to miscommunication and
incomplete information integration and flow.
·
In large organizations, some individuals assume the task of keeping historical
records of previous projects in idiosyncratic forms. These individuals become
the `librarians' of part of the organization's most critical information.
·
Different designers (and groups of designers) use different vocabularies to
describe the same or very closely related sets of things.
·
Individuals tend to organize information in ways understandable to them,
generally in the form of sketches and notes. There is a substantial overhead in
merging these different views to arrive at a consensus.
·
Tools that discover terminological differences can focus negotiations and
improve the use of information.
From these observations, we have
identified some basic features that we have found to be important in satisfying
the requirements elaborated earlier. In the following section we identify the
features and their role addressing the requirements through an example.
Computation and participation
Many manual techniques have been proposed
and used in the past, depending on the artifact and participation characteristics;
among others, they consists of workshops, manuals, educational material, games,
model building and ways of studying users/customers' needs [14, 1, 3, 5, 11, 16, 17].
Since there are some successful
participation projects carried out in computerless environments indicating that
computers are not a prerequisite for participation [1, 2, 3, 4,
5, 6,
7, 8, 9] the question of the need for tools for
supporting participation arises. The question is whether computational
techniques can extend the scope of participation. Some evidence that
computer tools can influence participation is obtained from studying the
patterns of interactions different people display via electronic mail or other
media systems [25]. For instance, some people who
did not participate in face-to-face discussions found it easier to participate
through an electronic medium.
Our ideal of participation attempts to
relax the limitations of participation. It enables all potentially affected by
a product to participate. This requires condensing data created in the course
of extended participation with tools instead of reducing the potential of
producing data by restricting participation. Furthermore, computer tools that
can transform a large amount of information into a comprehensible form may
support fast informed contextualized decisions [45].
The ideal participation allows potential participants to plan their
participation according to its duration; if some cannot or are not willing to
participate fully in all discussions, they may monitor the discussion coded
through computer medium and comment on them. Their comments are a form of
restricted participation but it provides input otherwise inaccessible. If some
participants stop participating due to other constraints, their original input
remains accessible later. Finally, one can envision potential participants who
are unable to participate due to conflicts in schedules-a situation common in
urban planning participation projects in small towns [30];
in these cases, tools which are designed to allow for asynchronous
participation can permit those potential participants to contribute [28]. While the local newspapers have served as a record
of such participation [30], yesterday's newspapers
are sometimes hard to come by.
The ideal participation treats all
participants as equal in a co-design or dialectical activity. The dissemination
of information in a format accessible to all participants can loosen
traditional hierarchical structures that increase opportunities for
participation; however, there are risks embedded in this scenario [46]. First, users may initially develop resistance to
programs that question their judgment. Second, users may develop a tendency to
rely on the tool rather than exercising their own judgment. Third, data
containing participation records can be used to monitor and control
users/designers. We argue that such risks can be minimized in the development
of computational tools by involving users in their development.
Different computer tools support the
communication that underlies participation to various degrees. Some tools provide
the basis for participants to communicate via a shared workspace [31]. Other tools support the communication process by
recording the structure of issues raised in the participation process (gIBIS,
Conklin and Begeman, 1988). And still other tools can elicit issues from
multiple participants, provide feedback and guidance, as well as perform
analysis to detect discrepancies between different views (KSS0, Gaines and
Shaw, 1989).
Whereas all the above tools were developed
for collaboration of experts familiar with the use of computers, no such
assumption about the proficiency of participants (customers as well as
designers) with computers can be made with respect to participation in domains
such as architectural design or urban planning. Therefore, in order to provide
computational support for participation we need to step back, look at the
evolution of this idea in the last two decades, and propose a technique that is
informed by past experience and recent empirical analyses of design processes.
Different computer tools may support
communication patterns useful for participation to various degrees. However,
all these tools were developed for collaboration of experts familiar with the
use of computers. No such assumption about the proficiency of participants
(customers as well as designers) with computers can be made with respect to
participation in domains such as architectural design or urban planning.
Therefore, in order to provide computational support for participation we need
to step back, look at the evolution of this idea in the last two decades, and
propose a technique that is informed by past experience and recent empirical
analyses of design processes.
The early visions about the role of
computers in participatory design [14] were
optimistic. One study discussed the development of an environment that would
have intelligence, common sense, and that would be able to learn and be
responsive [47]. Such an environment would
``participate'' on-line with people in a process leading to a better
environmental state. This idea conceals a critical difficulty. In order to
eliminate the need for professionals, an intelligent environment would be
built. This environment, however, would need to embed all the ``knowledge'' of
these professionals and would be developed by those who presumably can
appreciate all the issues in architectural design or urban planning a
priori. Instead of creating an open environment, this idea can lead to
imposing greater control than currently is exercised over a human built
environment.[x] Another study proposed to use
CAD tools with a participatory twist, namely, use CAD systems as a means of
storing common representations [48]. Other
studies attempted to quantify value judgment and use computers to help in
arriving at common quantifications in some manner [49].
All these studies assumed that the terminology used by different participants
was the same and that the objectives were shared by all--a manifestation of the
Platonic view of knowledge and design.
Since the 70's, there have been additional
attempts to provide computerized support for participation in the form of
design aids [50, 12]. To illustrate, PARTIAL was a layout tool
that allowed users to set-up general goals, define preference features,
manipulate graphical layout, and evaluate it [12].[xi]
PARTIAL was not really a tool for participation but simply a design aid, and
one that assumes the analytic part of design as the critical. In fact, PARTIAL
suffered from problems similar to those raised before about creating the
intelligent environment as commented by Watts and Hirst (1982, p. 18): ``It is
essential that design tools are congruent with the decision-making structure,
even if used as a vehicle for learning. No tool will be useful unless it is
congruent with the designer's definition of the problem.'' By implementing a
simple fixed procedure for layout design, the program assumed that it was
congruent with the way designer and users would solve layout problems.
Other tools that support some
participation include facilities for storing catalogues of designs to present
to potential customers [50] and facilities
for layout design of rooms, 3D graphical display of homes, especially when the
choices available are among modular units.
In the 1990's we are less sanguine and
more skeptical about: (1) the transfer of power regarding design decisions --
the problems surrounding participatory design are not a simple matter of a
universal inverting of the power structure between, for example, designers and
users; and (2) assuming the existence of sophisticated computer installations
for supporting design participation in the near future. We conjecture that what
is needed is a focus on specific design contexts, each of which exhibits its
own particular problems of interpretation and translation of varying user and
designer perspectives, and the honing of computer support tools in a participatory
atmosphere responsive to differing design circumstances. In this respect, we
also differ with more recent proposals for computer-supported participatory
design [52, 53].
There are several critical issues that a
system for supporting participation must address.
Usability. Participants in design projects will often lack experience
with computers. It is critical that computer tools be usable by participants.
Furthermore, as observed in many studies about the introduction of software
into organizations, such support tools may not be congruent with the way
participants think. In such cases, tools remain unused.[xii]
Our approach to the development of computer tools--participatory design and
evolutionary prototyping--is geared towards alleviating this problem [54].
Type of modeling. Empirical studies show that engineers use
a variety of modeling techniques (see summary in Subrahmanian, 1992). Therefore,
it is our contention that no single representation or abstraction technique can
be imposed on designers as well as other participants a priori,
without severely limiting their ability to model effectively. We thus use a
notion of conceptual information modeling that allows multiple classifications
to be imposed over a corpus of information. Abstraction levels are imposed by
the users, in whatever way they see fit.
Empirical studies and critical analyses of
modeling activities in engineering [36] also show
that the majority of design activities consists of informal, rather than
formal, modeling. Informal modeling is even more suitable for modeling
value-based, rather than technology-based, decisions. Consequently, a tool for
supporting participation must support informal modeling. This property should
not, nonetheless, compromise the ability to incorporate formal models as
participants see fit.
Extent of participation. If design is broadly interpreted, all
those potentially affected by a project are not known. This is due in part to
the lifetime of artifacts and to the inclusion of life-cycle concerns in
design. Therefore, in urban planning participation involves all future users of
buildings as well as users of existing buildings. Facilitating continuous
participation requires using innovative techniques. To illustrate, computer
tools accessible through terminals can collect feedback from users of existing
buildings.
A support for participation activities
over an extensive time period can be realized by addressing two critical
functionalities:(1) facilitating the creation and maintenance of shared
memory--an evolving corpus of shared information including its content and
meaning [27, 55] and
(2) providing for communication channels for creating this memory [28]. Since the role of communication has been briefly
discussed in Section 3
and in [28], we only elaborate further on the concept
of shared memory.
Shared memory is critical for design [27,
56]. It is critical for the evolution of a
discipline, for avoiding repeating errors, and for communication [57]. The difficult aspect is understanding how
shared memory is created, maintained, evolves, and used and how these
activities can be supported. As we said in our objection to the Platonic view,
shared memory cannot be constructed by individuals; rather, it is socially
constructed through negotiations and reconciliations, while maintaining the
differing perspectives as legitimate [55].
Those viewing architects or designers as
elite may chose to ignore the need or even the benefits from participation.
They can present forms of ``shared memory'' constructed by architects, such as
the pattern language [58], that can be used
for participation. Nevertheless, even they cannot deny the fact that in large
projects, several architects or designers operate together and must reconcile
their views through some communicative process. For simple examples, revealing
patterns of participation through a medium of shared memory among members of
design groups see:
(1) Peng (1992);
(2) a knowledge acquisition
study of tall building design illustrating the interdependence relationships
and communication processes between architects and structural engineers, see
Meyer and Fenves (1993);
(3) the use of mappings between
languages used to express designs so that the designs can be analyzed by
cluster analysis [61]; and
(4) the creation of a shared
memory by using a multi-exposure photograph of different models, all taken from
an identical viewpoint [56].
The photograph in the latter example
indicates the perceived similarity between the models from the particular
viewpoint. These similarities capture the shared design archetype of the group
whose members created the different models. Note that present technology of modeling
3D objects by computers can be used to obtain better fits by searching through
a set of potential similarity measures. Once shared memory is recognized as
critical to design, in general, and participatory design, specifically, its
creation and management needs to be addressed.
We can envision additional functionalities
incorporated in a computational support tool, all helping in the communication
process and thereby leading to the creation of a shared memory. For example,
full scale model building allows participants to evaluate designs in a way
significantly different from observing drawings or 3D scale models [1, 3]. In
the future, graphics technology including virtual reality techniques could
provide similar functionalities with fewer resources.[xiii]
Thus far, we have articulated various
dimensions of participation based on the experience of participatory design as
reported in the literature and our own observations of, and participation in,
design projects. These dimensions and their interpretations are bound to be
incomplete and approximate. Their further extension requires the collection of
information from many participation projects and its analysis. It is this function
that computational support tools introduce that are lacking in other
techniques. Computational support tools that are used for participation can
store the actions and products of users' participation and be used later for
analysis. The use of computational tools can remove the significant hurdle of
collecting and coding information. Note, however, that the analysis cannot be a
solitary activity of researchers, but again, a participatory activity with the
other participants (for more details on participatory action research see
Reason (1988) and Whyte (1991)).
We now describe n-dim, a computer
tool that is designed to facilitate modeling starting from the initiation of a
design process and continuing throughout the life-cycle of the artifact [34, 35]. It turns out
that the functionality designed into n-dim matches the functionality
required from computational tools for participation.
n-dim
We begin by introducing the building blocks
of n-dim--objects and models. There is a basic cleavage in the space
of n-dim objects between atomic and structured
objects. As the name indicates, atomic objects cannot be broken down
any further, e.g. an integer, a link, a piece of text, an image, an audio
bitstream, etc. One could think of atomic objects as things that have values
of some sort.
The primary form of structured
object is the model. A model is a set of links, which are, themselves, atomic
objects. Link types are given their meaning(s) by the modeling language(s) in
which they occur. It is quite possible to have the same link type mean totally
different things in different contexts; we view the meaning of links as
something to be negotiated by users of the system over time.
Operationalizing the semantics of particular interpretations of links is
considered an open-ended process; n-dim provides mechanisms for doing
so, but does not require it to be done in order to use a link type. All
objects, whether structured or not, are constructed using another model as
their modeling language. Typically, modeling languages specify what
objects can be in a model and what relations they can have to one another. Such
specifications can be thought of as grammars.
To illustrate these concepts, Figure 1 shows the model Should_I_Participate created by an n-dim user with the Universal modeling language. The model shows that
the object labeled Self_Interest
influences the object labeled Participation, Participation requires Resources, Participation influences the Quality_of_Design, and Time is a Resource. The legal types of objects and links in the Universal modeling language include all existing
types known to n-dim, as well as additional arbitrary types defined by
the user.
Figure 1: A simple n-dim model
There is a special link called a part link, which is canonically represented as
a box inside of a box. That is, Should_I_Participate in Figure 1 contains five part links, all whose source is Should_I_Participate itself and whose targets are the objects
displayed ``inside'' the model. It must be stressed that, Should_I_Participate does not contain any of the five
objects that appear inside it; this is simply the canonical n-dim
visual representation for a part link. The four other links in Should_I_Participate are represented as directed lines, which is the
normal representation of all types of links except for the part link.
In Figure 1, the
object Participation is itself a model. One can view its
content by ``opening'' it. The process of traversing models through opening
them (i.e., tracing part links) is a browsing activity. One can also browse the data in n-dim
models in different ways (e.g., through directed search), depending on the
models one chooses to traverse.
One can map a structure of an n-dim
model onto multiple projections, which discriminate between possible
views of that structure. Finally, any projection can be mapped onto multiple presentations,
which fix the characteristics of projection vis a vis its rendering.[xiv]
n-dim uses the same representational mechanisms to deal with all three
levels, projections are models, as are renderings. The system merely interprets
such models appropriately when needed.
n-dim models can play at least two different roles
for an n-dim user: instance/prototype and language. An n-dim
model is in a specific context. When many contexts are found where such models
are useful, a modeling language, that is, a model describing the possible
structure of other models, can be created.
To illustrate, Figure 2
depicts two additional models created in the Universal modeling language, where node and link
types can be any known to n-dim. Figure 3 shows a
model created from the models in Figures 1 and 2 in a new language, IBIS (issue-based language), that may have evolved
from observing similarities in the previous models and the wish to integrate
them.
Figure 2: Two additional models
Figure 3: Transformation into a new
model
The language role is easier to describe. A
language has in mind some restrictive purpose and functionality. Whereas the Universal language, used to construct the models in
Figures 1 and 2, allows the use of
arbitrary nodes and links, the languages MLTransformation and IBIS specify a small number of legal types of nodes
and links that have rather transparent meanings. The more restrictive or
strongly typed the language, the more amenable it is to the applications of
formal methods. As a simple illustration, in the Issues_in_my_Participation model, the structure of sub-issues-of
could be used to determine when the goal issue Decide has been attained. Or, if issues would be
restricted (or specialized) to contain numeric information about the expected
duration to solve them, the structure of issues could be used to find the
critical path to address the goal issue.
The desire to formalize or standardize as
a primary objective often leads to (1) the creation of formal techniques for
design such as grammars [62, 63], (2) the establishment of standards such
as STEP, or (3) the proposal of various formal data models such as EDM [64] or others [65]. n-dim
is developed to provide support for modeling all these endeavors including the
embedding of tools such as the layout design system ABLOOS [51, 34]. In fact, such
modeling activities are relatively easy, albeit quite involved, compared to
providing support for informal modeling such as sketching, communicating, or in
general, to the social construction of shared memory.
The prototype and language roles of n-dim
models play a crucial part in the usability of n-dim in the sense that
they allow users to mix bottom-up and top-down processes. In the former, a user
creates models of specific situations and from these generalizes to create a
modeling language as demonstrated in Figures 2 and 3. In the latter, the language is defined from which
individual instances of the model can be created and elaborated to use rules
(see below) and other external computational agents.
While a modeling language defines the set
of possible instances of that language, it does not necessarily define the set
of meaningful instances. Some constraints can easily be imposed on a
language for conveying simple semantic information. For example, in the IBIS modeling language, the link sub-issue-of
can link two ISSUE objects. To
capture more complex semantic, as well as syntactic, information about models,
additional facilities are needed. The ability to put rules in modeling
languages provides these facilities [34].
n-dim contains additional facilities that are
critical for the participation of multiple people in a design project. They
deal with the creation, communication, and analysis of information. These
categories overlap and feed one into the other; they include Publication,
Search and the Talk facilities, and partially integrated
natural language processing and machine learning capabilities. We briefly
discuss each of them in turn.
Publication (creation facility). Participation is a situation
where the personal and the community intermingle; therefore, n-dim
must allow for maintaining personal and community workspaces. In the personal
workspace, participants can create their models in privacy. Once participants
decide to share models, they ``publish'' them, thereby allowing all the
remaining participants to inspect them. Published objects are persistent, they
remain unchanged. Only their copies can be altered. Published objects provide the
means for keeping design history and facilitate the creation of shared memory.
To illustrate, Figure 1 depicts a published object. It is
distinguished from ordinary objects by the two stones mark to the left of the
model's name.
Talk (communication facility). The talk facility
provides a means for synchronous communication. This facility allows
participants to exchange short notes or arbitrarily complex information such as
models.
Search (communication and analysis facility). Publishing
and searching are critical activities in modeling by multiple participants.
They provide the basic substance for asynchronous communication. Participation
can easily generate an enormous amount of information in the form of published
models. As the corpus of information increases continually, facilities are
needed to search it for information relevant to current modeling activities.
Currently, search is performed via a structured query editor. n-dim
allows for, and will include, techniques designed to facilitate search by
models, whereby a user could specify a partial model and search for models
closely related to the outlined model. These techniques would be significantly
more usable to computer illiterate participants.
Natural language processing and machine
learning (analysis and
creation facilities). Natural language processing will allow participants to
discover terminological patterns implicit in large text corpora written by
previous participants. The discovered patterns can act as a basis in building
conceptual models thus aiding in the creation process. Further, such patterns
can be used to create synonyms for extending search queries beyond
idiosyncratic labeling. The machine learning capability will allow participants
to (1) inspect large amount of technical information and transform it into
comprehensible forms; (2) learn and evolve the understanding of the current
design problem through analyzing its progress and its relation to previous
designs; and (3) compile information for future reuse.
All the above facilities are expected to
be executed under participants' control; there is no assumption that any of
them will be executed automatically.
Example
This section discusses a hypothetical
example of using n-dim for facilitating design participation in the
design of the new community library in a hypothetical town:
The first stage in any participation is
the initial presentation of concepts that is intended to trigger feedback or
dialogue. In this project, the trigger was a town meeting in which the new
project was presented to the town assembly. The county administrators decided
to try a new approach in designing the facility: participatory design. They
have heard a little about it and thought that it may be interesting and also
beneficial.
The administrators had to discuss many
issues: what could be the form of the participation, will they need a
moderator, how long will the process take? They had to collect material on
workshops and manuals for participation. Although some information was
gathered, it was still partial.
Meanwhile, the county administrators found
a computer program called n-dim that was available for experimenting;
although it was not commercial quality software, it was free and was
accompanied with a proposal to send a researcher interested in studying participation
to act as a participant in the project.
In the town meeting, the county officials
presented the new idea and asked several people moderately proficient with the
use of personal computers to try the software, and meanwhile, learn about
participation by using it. The county officials also scheduled another meeting
to allow these residents report their experience with the software. Moreover,
this second meeting was supposed to give residents the ability to hear about
the tool and the concept of participation from members of the community, rather
than from county officials, and subsequently decide on the future progress of
the project.
The researcher had several roles that
facilitated the understanding of participation. Most importantly, the researcher
demonstrated by example that n-dim could be used to structure
information, besides being used as a fancy electronic mail facility. Figure 4 shows a text written by Joe and its translation to the ToDo model by the researcher. Meanwhile,
Keren, another active resident, created the models in Figures 2
and 3, among others.
Figure 4: From textual to model
representation
In order for Joe to view the ToDo model the researcher created, the
researcher had to publish it, and Joe had to search for it (or receive it
through the talk facility if the researcher was using n-dim at that
time). Joe searched for the published ToDo objects and also found Keren's model which she
recently published. Joe could then view and use his model in various ways; for
example, he could annotate it and send it to Keren or copy and use some of
Keren's model in his ToDo model.
Once the residents understood the
modeling, browsing, search, and talk facilities of n-dim, they could
start learning about previous experiences other participants had while
participating in different projects and exchange their impressions. Such
contextual information was critical for better appreciating the meaning of
objects such as Frame_of_Mind, Resources, or Quality_of_Design appearing in Figure 2.
Therefore, in order to make an informed decision about whether to participate,
residents had to understand the issues involved in the library project, even if
in a preliminary manner.
Figure 5 shows the
model Participation_Projects classifying previous participatory design
projects in addition to the model Facilities_Classification classifying facilities according to their
function. These models were created by the researchers before the beginning of
the exercise, but in real use, they are expected to accumulate through actual
use in design projects. The participants detected the Boulder_Creek_Branch_Library and Library models and decided to browse the latter. The
participants also noticed the models labeled
Figure 6 shows four
models explored by the residents: (1) the Library model organizing the information on library
design; (2) the Experience model
containing the details about specific projects and their relation to specific
design stages; (3) the Guidelines model containing various information sources about library design; and (4)
the Floor_size model depicting guidelines for
determining areas of various spaces of a library.[xv]
Figure 7 shows some of the issues in library
design distilled from the different experiences and guidelines.
Figure 5: Browsing previous
participation experience
Figure 7: Issues in designing a
library
One of the cases appearing in the Experience model, the Boulder_Creek_Branch_Library model, is an example of designing a
community library center [4]. The
residents decided to explore it next. In this model, the residents could view
stages in the design process, including layouts of proposed solutions (see
Figure 8). The residents could study this design and
appreciate its relevance to their project. Indeed, a close observation revealed
interesting similarities with their project. The success of the Boulder Creek
Library design was influential on deciding to participate in the library
design. The residents also discovered that the nature of the participation
process, even if carefully planned ahead, could change as it unfolds.
Therefore, their decision does not bind them to specific activities or
processes; rather, it only requires that they maintain an open Frame_of_Mind.
Figure 8: Browsing the Boulder
Library experience (drawing after [4])
The decision was to adopt participation
and try n-dim as a vehicle for its facilitation in the project. The
next steps for the participants included among others: continuing the process
of idea elicitation, requirement formulation, consensus establishment through
building models in n-dim and communicating them; contacting potential
architects and constructors that were involved in participation or community
library design projects to propose their participation in the early stages of
the project; and possibly using and viewing a variety of analysis and design
tools relevent to library design.
As mentioned in the discussion of n-dim,
the embedding and use of design tools within n-dim is an important
concern of the n-dim research project and is central to the goals of
participation in design. By creating various kinds of models, participants are
able to discuss, annotate and provide feedback on the results of using design
tools to those who generated the results. At the least this would enable
participants at every level to access, review and monitor the tool based design
activities of design professionals in a direct way that is normally not
possible today. Depending on the sophistication of the design tools involved
and their ease of use within n-dim, participants may also be enabled
to use certain tools themselves. Special n-dim modeling languages can
assist in automatically preparing input to the tool in the proper format and sequence,
in translating the results of the tool's execution back into n-dim
models, and in directing input and output between tools (this is the common
case where the output of one tool is filtered to provide input for another, and
so on).
In Figure 9, we
illustrate the use of a layout design tool called ABLOOS to produce layouts of
the library. The residents used the spaces in the Boulder Library project as
templates to be configured by some of the guidelines found in the Guidelines model, for exploring a preliminary design
for their library. The figure shows, in the upper left, two alternative Library_layout_Input models. These were constructed based on
templates for the various library spaces contained in a Library_Design_Repository model - this model is a classification of
the spaces generally included in libraries such as those in the Boulder Library
project. The Library_Layout_Input1 model is shown open to display the desired spaces and their adjacency
relationships. A model of one of the spaces, Reference, is shown open to illustrate its
recommended maximum and minimum sizes and minimum area - this can be edited and
adjusted for its instantiation in a particular library layout. In the upper
right, the Run_Library_Layout
model illustrates the Library_Layout_input1 model linked as input to a Run_ABLOOS model to generate layouts as output. The
output is stored in a Layout_Results model and the Run_ABLOOS model is
a standard, pre-existing model available to the participants as a result of the
embedding of ABLOOS as a tool within n-dim. The open Layout_Results model shows two alternative layout models
for the library generated by ABLOOS and these layouts are displayed in the
bottom of the figure.[xvi] These solution models can now be
included in (pointed at) from other models where discussions and annotations
about their merit can be attached and communicated to the library designers and
other participants. For instance, the solutions might be referenced in an IBIS
issue-base discussion model.
Figure 9: Conducting a preliminary
design with ABLOOS
Meanwhile, while the design was
progressing, the researcher was learning about participatory design and the
(often creative) ways n-dim was used by the participants in the
project. The researcher heard feedback from participants and attempted, with
assistance from the n-dim group, to provide timely answers. The
researcher also participated in the design itself by assuming a role assigned
by all the participants. This provided the researcher the necessary baseline
for understanding the issues raised in the project, the terminology used, and
the roles different participants assumed. Such information was modeled
continually in n-dim and provided further details for analyzing the
progress of the participation process. The information on this project as it
evolved was modeled and inserted in its proper place into the models of past
participation experiences. This activity was performed in the everyday course
of the project.
Summary and future work
Participation in design is a primary right
of all potentially affected by a design project. We have discussed some
obstacles to participation and proposed computational support systems as
vehicles for extending the scope of participation as it is practiced, when it
is practiced. We have concentrated on some issues that pertain to architectural
design and urban planning and their implication to the design of computational
support tools.
Our approach is radically different from
early ideas about computational support of participation which might be
described as a means for the elimination of experts from design. Our approach
is based on providing participants with communication facilities and
information about past experiences that they can themselves explore and
organize. Our approach is meant to facilitate learning and growth. We have
presented a computer tool n-dim that integrates facilities ranging
from unstructured to structured representation of information, communication,
and design tools integration, and illustrated its use in a hypothetical design
case.
In the example, we have demonstrated how n-dim
can be used to educate prospective participants about participation. Such
education consists of explaining the conceptual issues involved in
participation and the process of participation, in addition to providing
background on projects relevant to the project presently conceived. We have
briefly noted how a collaboration can take place between participants to arrive
at consensus during participation. This collaboration can involve diverse
aspects such as discussing the issues of the preliminary design or designing
the layout of the building with the aid of computational tools.
There are evidently many issues that need
to be addressed in future (participatory) research; they all relate to the
management of vast amount of information and its usability for future
participants. We plan to tackle these issues and others that are uncovered
through participatory activities in which n-dim is used in design
activities. The information accumulated from participation projects using n-dim
and participants' feedback could be used to study and further understand
participation. In the long run, evidence that supports the utility of n-dim
in facilitating participation can only come from using it in participation
projects. This, in turn, would give feedback and insight on the further
development of n-dim as a tool for participation.
From the perspective of developing
computer tools, participation serves three purposes. First, it is required for
developing a usable tool. Second, it is needed for testing the tool. Third, it
is required for collecting data on people participating in various design
scenarios, thereby gaining a better understanding on the issues involved.
Furthermore, based on this view, we have provided the beginnings of a
specification for computer tools to support participation, including briefly
outlining the functionality of its prototype.
We are currently pursuing several projects
with industrial affiliates to test these ideas following the deployment of n-dim
in these affiliates' sites. In addition, in the fall of 1993 n-dim
will be used in a senior-level software engineering course which features a
team-based approach driven by the use-case methodology developed by Jacobson [69]. With our affiliates, we intend to study and
model the evolution of n-dim as it takes place through the cycles of
implementation, use, and evaluation. We (now n-dim developers as well
as users) can then trace this evolution to its sources in different
participatory activities, thereby studying participation as we design.
Acknowledgments
This research has been supported in part
by the
References
G. Hardie, ``Community participation based on three-dimensional
simulation models,'' Design Studies, vol. 9, no. 1,
pp. 56-61, 1988.
R. Kaplan, ``Citizen Participation in the design and evaluation of a
park,'' Environment and Behavior, vol. 12, pp. 494-507,
1980.
R. J. Lawrence, ``Simulation models as a medium for participatory
design,'' in Design: Science: Method, Proceedings of The 1980 Design
Research Society Conference, (
J. Oberdorfer, ``Community participation in the design of the Boulder
Creek Branch Library,'' Design Studies, vol. 9, no. 1,
pp. 4-13, 1988.
H. Sanoff, ed., Designing with community participation.
H. Sanoff, ed., Participatory Design: Theory and Techniques.
W. F. Whyte, ed., Participatory Action Research.
W. F. Whyte, D. J. Greenwood, and P. Lazes, ``Participatory
action research: Through practice to science in social research,'' in Participatory
Action Research (W. F. Whyte, ed.), (
W. F. Whyte and K. K. Whyte, Making Mondragon: The Growth and
Dynamics of The Worker Cooperative Complex.
W. G. Adams, ``Participatory programming for Digital Equipment
Corporation, Inc,'' Design Studies, vol. 9, no. 1,
pp. 14-24, 1988.
H. Sanoff, Visual Research Methods in Design.
J. Watts and M. Hirst, ``User participation in the early stages
of building design,'' Design Studies, vol. 3, no. 1,
pp. 11-18, 1982.
G. Bylinsky, ``The marvels of `virtual reality.','' Fortune,
vol. 123, p. p138(5), 1991.
N. Cross, ed., Design Participation.
S. Frary and J. Seelen, Cherokee,
R. Sommer, Social Design: Creating Buildings with People in Mind.
J. Q. Wilson, ``Planning and politics: Citizen participation in urban
renewal,'' Journal of the American Institute of Planners,
vol. XXIX, no. 4, pp. 242-249, 1963.
J. Nielsen, Hypertext and Hypermedia.
A. Oliveira, ed., Hypermedia Courseware: Structures of
Communication and Intelligent Help.
J. F. Nunamaker, L. M. Applegate, and B. R. Konsynski,
``Computer-aided deliberation: Model management and group decision support,'' Operations
Research, vol. 36, no. 6, pp. 826-848, 1988.
J. F. Nunamaker, A. R. Dennis, J. S. Valacich, D. R.
Vogel, and J. F. George, ``Electronic meeting systems to support group
work,'' Communications of The ACM, vol. 34, no. 7,
pp. 40-61, 1991.
A. Pinsonneault and K. L. Kraemer, ``The effects of electronic
meetings on group processes and outcomes: An assessment of the empirical
research,'' European Journal of Operational Research, vol. 46,
no. 2, pp. 143-161, 1990.
D.
P. Reason, ed., Human Inquiry in Action: Developments in New
Paradigm Research.
L. Sproull and S. Kiesler, ``Computers, networks and work,'' Scientific
American, vol. 265, no. 3, pp. 116-123, 1991.
J. Balasubramanian, ``Influence of status on group interaction
processes: electronic mail versus face-to-face discussions,'' in Proceedings
of the Human Factor Society 31st Annual Meeting: Rising to New Heights with
Technology, Vol 2, (Santa Monica, CA), pp. 1355-1359, Human Factors
Society, 1987.
S. Konda, I. Monarch, P. Sargent, and E. Subrahmanian,
``Shared memory in design: A unifying theme for research and practice,'' Research
in Engineering Design, vol. 4, no. 1, pp. 23-42, 1992.
E. Subrahmanian, R. Coyne, S. L. Konda, S. N. Levy,
R. Martin, I. A. Monarch, Y. Reich, and A. W. Westerberg,
``Support system for different-time different-place collaboration for
concurrent engineering,'' in Proceedings of the 2nd IEEE Workshop on
Enabling Technologies Infrastructure for Collaborative Enterprises (WET ICE),
(Los Alamitos, CA), pp. 187-191, IEEE Computer Society Press, 1993.
S. Minneman and J. C. Tang, ``VideoWhiteboard: Video shows to
support remote collaboration,'' in Proceedings of the Conference on Human
Factors in Computing Systems, (
H. Sanoff, ``Participatory design in focus,'' Architecture and
Behavior, vol. 4, no. 1, pp. 27-42, 1988.
H. Ishii and N. Miyake, ``Toward an open shared workspace:
Computer and video fusion approach of team workstation,'' Communication of
The ACM, vol. 34, no. 12, pp. 37-50, 1991.
J. Conklin and M. L. Begeman, ``gIBIS: A hypertext tool for
exploratory policy discussion,'' ACM Transaction on Office Information
Systems, vol. 6, no. 4, pp. 303-331, 1988.
B. R. Gaines and M. L. G. Shaw, ``Comparing the conceptual
systems of experts,'' in Proceedings of The Eleventh International Joint
Conference on Artificial Intelligence, (Detroit, MI), pp. 633-638,
Morgan Kaufmann, 1989.
S. Levy, E. Subrahmanian, S. L. Konda, R. F. Coyne,
A. W. Westerberg, and Y. Reich, ``An overview of the / environment,''
Tech. Rep. EDRC-05-65-93, Engineering Design Research Center, Carnegie Mellon
University, Pittsburgh, PA, 1993.
E. Subrahmanian, A. W. Westerberg, and G. Podnar, ``Towards
a shared information environment for engineering design,'' in Computer-Aided
Cooperative Product Development, MIT-JSME Workshop (Nov., 1989)
(D. Sriram, R. Logcher, and S. Hukuda, eds.), (
E. Subrahmanian, S. L. Konda, S. N. Levy, Y. Reich,
A. W. Westerberg, and I. A. Monarch, ``Equations aren't enough:
Informal modeling in design,'' Artificial Intelligence in Engineering Design,
Analysis, and Manufacturing, vol. 7, no. 4, pp. 257-274,
1993.
L. L. Bucciarelli, ``An ethnographic perspective on engineering
design,'' Design Studies, vol. 9, no. 3, pp. 159-168,
1988.
S. Hales, Analysis of The Engineering Design Process in an
Industrial Context. PhD thesis, Department of Engineering,
S. Finger, E. Subrahmanian, and E. Gardner, ``A case study
in concurrent engineering for transformer design,'' in Proceedings of
ICED-93 (The Haugue) (N. F. M. Rosenburg, ed.), (
H. Petroski, ``Paradigms for human error in design,'' in Proceedings
of the 1991 NSF Design and Manufacturing Systems Conference (
P. Sargent, E. Subrahmanian, M. Downs, R. Greene, and
D. Rishel, ``Materials' information and conceptual data modeling,'' in Computerization
and Networking of Materials Databases: Third Volume, ASTM STP 1140
(T. I. Barry and K. W. Reynard, eds.), American Society For Testing
and Materials, 1992.
E. Subrahmanian, ``Notes on empirical studies of engineering tasks and
environments, invited position paper,'' in NSF Workshop on Information
Capture and Access in Engineering Design Environments (Ithaca, NY),
pp. 567-578, 1992.
A. Westerberg, M. Rychener, P. Piela, and L. Biegler,
``White paper on engineering design.'' Report to Westinghouse Corp.,
Engineering Design Research Center, Carnegie Mellon University, Pittsburgh, PA,
1988.
D. J. Wilkins, J. M. Henshaw, S. H. Munson-Mcgee, J. J.
Solberg, J. A. Heim, J. Moore, A. Westerberg,
E. Subrahmanian, L. Gursoz, R. A. Miller, and G. Glozer,
``CINERG: A design discovery experiment,'' in NSF Engineering Research
Conference, (Amherst, MA), pp. 161-182, College of Engineering,
University of Massachusetts, 1989.
Y. Reich, S. Konda, S. N. Levy, I. Monarch, and
E. Subrahmanian, ``New roles for machine learning in design,'' Artificial
Intelligence in Engineering, vol. 8, no. 3, pp. 165-181,
1993.
S. Zuboff, In The Age of The Smart Machine: The Future of Work and
Power.
N. Negroponte, ``Aspects of living in an architecture machine,'' in Design
Participation (N. Cross, ed.), (
W. Mitchell, ``Experiments with participation-oriented computer
systems,'' in Design Participation (N. Cross, ed.), (
T. Maver, ``Simulation and solution teams in architectural design,''
in Design Participation (N. Cross, ed.), (
J. C. Carp, ``Design participation: New roles, new tools,'' Design
Studies, vol. 7, no. 3, pp. 125-132, 1986.
R. F. Coyne, ABLOOS: A Computational Design Framework For Layout
Synthesis. PhD thesis, Department of Architecture,
C. M. Duursma, ``Support for understanding and participation in a
distributed problem solving system,'' in Human-Computer Interaction.
INTERACT '90. Proceedings of the IFIP TC 13 Third International Conference
(D. Diaper, D.and Gilmore, G. Cockton, and B. Shackel,
eds.), (
M. Quayle, ``Computer-aided participatory design: SITESEE,'' Journal
of Architecture and Planning Research, vol. 4, no. 4,
pp. 335-342, 1987.
Y. Reich, S. Konda, I. Monarch, and E. Subrahmanian,
``Participation and design: An extended view,'' in PDC'92: Proceedings of
the Participatory Design Conference (
I. A. Monarch, S. L. Konda, S. N. Levy, Y. Reich,
E. Subrahmanian, and C. Ulrich, ``Shared memory in design: Theory and
practice,'' in Proceedings of the Invitational Workshop on Social Science
Research, Technical Systems and Cooperative Work (Paris, France), (Paris,
France), pp. 227-241, Départment Sciences Humaines et Sociales, CNRS,
1993.
A. Ward, ``Design archetypes from group processes,'' Design
Studies, vol. 8, no. 3, pp. 157-169, 1987.
H. Petroski, To Engineer is Human.
C. Alexander, S. Ishikawa, M. Silverstein, M. Jacobson,
C. Peng, ``Participatory architectural modeling: Common images and
distributed design developments,'' in PDC'92: Proceedings of The
Participatory Design Conference (
S. Meyer and S. J. Fenves, ``Structural design of tall buildings,
knowledge acquisition study report,'' Tech. Rep. EDRC 12-58-93,
G. Best, ``Method and intention in architectural design,'' in Design
Methods in Architecture (G. Broadbent and A. Ward, eds.), (
G. Stiny, Pictorial and Formal Aspects of Shape and Shape Grammars.
U. Flemming, R. Coyne, T. Glavin, and M. Rychener, ``A
generative expert system for the design of building layouts - version 2,'' in Proceedings
of The Third International Conference on Artificial Intelligence in
Engineering: Vol. Design (
C. M. Eastman, ``Use of data modeling in the conceptual structuring of
design problems,'' in CAAD Futures '91: Proceedings (G. N.
Schmitt, ed.), (
M. K. Zamanian and S. J. Fenves, ``A framework for modeling and
communicating abstractions of constructed facilities,'' in CAAD Futures
'91: Proceedings (G. N. Schmitt, ed.), (
Public Library Association, Interim Standards for Small Public
Libraries.
R. Myller, Design of the Small Public Library. Bowker, 1966.
A. B. Hill, ``The small public library: Design guide, site selection,
design,'' Tech. Rep. R80-3, Center for Architecture and Urban Planning
Research, University of Wisconsin-Milwaukee, Milwaukee, WI, 1980.
I. Jacobson, M. Christerson, P. Jonsson, and
G. Overgaard, Object-oriented Software Engineering - A Use Case Driven
Approach.
Last
modified: Fri Oct 31 12:40:02 IST 1997