Generative Systems

Cities can be considered as complex information systems. They are adaptable, ever changing entities, which cannot be grasped through basic assumptions of how modernism used to do to plan cities. It processes its own information in an heterarchic way that defines patterns in an emergent behavior. Water, in this sense is a very crucial natural entity in the life of cities. However, it is commonly deleterious from the constructed environment, killing the ecologies it creates. Today, most of these water lands are diminished in size, either squeezed into canals or drained to open up space for streets. This can be defined as a wicked problem, since water as a natural entity always tend to find its own way sometimes by damaging the urban environment. Crucial point is to create the ecology with urban and natural environments again as a mutual understanding for a resilient city. This paper aims to search for how the ecological relationships of water in urban environment be reconsidered via computational simulation methods, while keeping its natural structure by adapting itself and manipulating the urban environment.

Seeking to leverage Expressive Range Analysis (ERA)-a method of characterizing generative systems-for analysis of generative AI (GAI) systems and their outputs, this paper categorizes the approaches that have been used for ERA. We present a taxonomy with five axes of characterization that may be applied to ERA methodologies: content agnostic vs semantic; quantitative vs qualitative; product vs process; objective vs subjective; and automated vs manual. While ERA has traditionally been limited to the domain of Procedural Content Generation (PCG) in video game development, we recognize parallels between PCG and GAI and hope to see an expansion of the application of ERA into the domain of GAI. Serving this goal, these axes provide metrics through which to categorize, compare, and explore approaches.

In today’s world of advanced artificial intelligence and large language models, it’s becoming nearly impossible to tell who truly created a piece of writing, code, or art. Old methods for proving authorship - like watermarks or file metadata - don’t work when content is constantly remixed and regenerated by AI.

The Recursive Anchor Protocol™ (RAP) is a new system that solves this problem by embedding special patterns, phrases, and structures directly into the work itself. These “anchors” make it possible for both people and AI to trace content back to its original creator, even after it’s been transformed many times. RAP offers a future-proof way to protect credit, ensure authenticity, and uphold intellectual rights in the age of generative AI.

Eylül 2008 Emre Cestel iii İÇİNDEKİLER KISALTMALAR v ŞEKİL LİSTESİ vi TABLO LİSTESİ ix ÖZET x SUMMARY xi 1. GİRİŞ 4.2. Yüksek Yapıların Kavramsal Tasarımında Üretken Yaklaşımlar 4.2.1. Doğadaki biçimsel benzetimlerin tasarım sürecinde kullanımı 4.2.2. Doğadaki süreç benzetimlerinin yüksek yapıların kavramsal tasarımında kullanımı 4.2.3. Yüksek yapıların kavramsal tasarımında parametrik ve kural tabanlı yaklaşımlar 5. YÜKSEK YAPILARIN KAVRAMSAL TASARIM SÜRECİNDE KULLANILACAK ÜRETKEN MODEL 5.

Understanding the cognitive processes that occur when designing with computational tools and methodologies is crucial as these new contemporary means of design will eventually be changing our designerly ways of thinking, necessitating a fresh look at design education to incorporate these evolving practices. From this point of view, this study aims to discuss generative design techniques in terms of design cognition. To do so, five generative systems that are commonly used in design disciplines are reviewed as Shape Grammars, Lindenmayer Systems, Cellular Automata, Agent-Based Systems, and Genetic Algorithms. The study explains the rules of each generative system, how they generate design alternatives, and how design domains utilize these systems based on the literature. The two main traditional approaches in the field of design cognition are focused in the paper as: design as a reflective practice and design as a problemsolving activity. Then, the study contemplates on the cognitive processes from the early stages of designing with generative systems to its end. Finally, an integrated cognitive approach regarding the use of generative systems in design exploration is proposed in the study as: design as a reflective problem-solving activity. In contrast to common uses of generative systems as result-oriented systems, this approach emphasizes these systems' potential in the early phases of design exploration; hence, highlights their potential in design practice and education.

Understanding the cognitive processes that occur when designing with computational tools and methodologies is crucial as these new contemporary means of design will eventually be changing our designerly ways of thinking, necessitating a fresh look at design education to incorporate these evolving practices. From this point of view, this study aims to discuss generative design techniques in terms of design cognition. To do so, five generative systems that are commonly used in design disciplines are reviewed as Shape Grammars, Lindenmayer Systems, Cellular Automata, Agent-Based Systems, and Genetic Algorithms. The study explains the rules of each generative system, how they generate design alternatives, and how design domains utilize these systems based on the literature. The two main traditional approaches in the field of design cognition are focused in the paper as: design as a reflective practice and design as a problem-solving activity. Then, the study contemplates on the cognitive processes from the early stages of designing with generative systems to its end. Finally, an integrated cognitive approach regarding the use of generative systems in design exploration is proposed in the study as: design as a reflective problem-solving activity. In contrast to common uses of generative systems as result-oriented systems, this approach emphasizes these systems’ potential in the early phases of design exploration; hence, it highlights their potential in design practice and education.

This study offers a productive system with a rule-based grammar
for original joint details of particular historical timber buildings. Traditional
construction techniques can be considered as both tangible and intangible
architectural heritage. We examined the structural formations of traditional timber joint details and determined shape rules according to the geometric arrangements of the shapes. These arrangements and basic spatial movements (rotation, translation, reflection) were created using the SortalGI plug-in code system. Depending on the formal rules in this system, a grammar has been created to transform decisions for restorative or replicating designs. In this way, optional outputs were obtained for the joint details of the timber element, which was digitised using the photogrammetry method. The possible outputs obtained in the digital space provide an accessible, predictable, flexible and interdisciplinary working environment for design processes. As a result, alternative output models were obtained as in digital space.

The population increase that occurred in North African and Middle-Eastern cities, has led to an uncontrolled urban growth that produced urban environments lacking the spatial richness from their historical vernacular towns. In order to alert and to change this context this paper illustrates work in progress for a design project at the city of Marrakech in Morocco. It describes from a historical and analytical point of view, research carried out at the Zaouiat Lakhdar quarter of the Marrakech Medina. The goal is to create the basis for a computational tool that could capture spatial features of the existing architectonic fabric and apply them at contemporary architectural scenarios. The methodology used is explained from the initial historical analysis and fieldwork realized in situ to the identification of social, religious and geometric set of laws and its codification into two different types of generative rules. These were than applied either at the urban level, or at the domesti...

This paper describes research carried out to develop a parametric urban shape grammar for the Zaouiat Lakhdar quarter of the Medina of Marrakech in Morocco. The goal is to create the basis for a system that could capture some features of the existing urban fabric and apply them in contemporary urban planning and architectural design. The methodology used is described, from the initial historical analysis and fieldwork to the identification of three subgrammars necessary to encode the complexity of the urban preexistences: the urban grammar, the negotiation grammar, and the housing grammar. Top-down and bottom-up approaches to grammar design are analyzed and compared. The bottom-up urban grammar developed is then described, and a hand derivation of the existing urban fabric is proposed. Visual, symbolic, and tagged computer implementations of shape grammars are briefly discussed and a novel design generated by the tagged interpreter is presented.

Visibility analyses models and tools predicting human perception have been developed in past years. The Spatial Openness Index, a visibility analysis model defined as the volume of the visible part of a surrounding sphere: the potential of a view and exposure that can indicate the Perceived Density. Visual Openness is one of the main objectives in the development of dense urban areas. This can influence the resident's satisfaction and the real estate value. Visual Exposure, referring to privacy aspects, is also a major aspect influencing the human environmental quality. Visual openness to a view and Visual Exposure expressing visual privacy are two leading, contradicting, but also complementary visual attributes developed as visual analysis for sustainable urban systems. Since different internal functions and activities demand different levels of privacy and can enjoy differently openness to the view, studying the internal space layout and functionality and looking at alternative design in regard to external space analysis is demanded. All models refer to geometrical attributes measured from the external façade towards the surrounding environment and all have an impact on the perceived density and other human environmental comfort, and may contribute to the planning and design process. In this paper, the above models are introduced and a study of the relation between internal space layout and functionality and external visual analysis is suggested.

This study describes the morphological urban order underlying Portuguese treatises and Portuguese urban cartographic representation produced from 16th century to 18th century. The historical documentation suggests that Pythagorean-Euclidian geometry appears to be a crucial ingredient for the understanding of Portuguese urban designthinking and urban design-making. To unveil the genesis of the morphological urban order present in the Portuguese colonial plans of the eighteenth century, a descriptive method, Shape Grammar has been adopted. Shape Grammar, as method, supports the analysis of the form-making logic and has proved to be powerful in shape analysis, description, interpretation, classification, evaluation and generation of a design language.

It deals with the critical concepts of Interoperability and Generativity and how they complement each other. Both are composed of several layers entailing different types of interconnections and linkages that have made the current technological boom possible.

The material is based on the work of Jonathan Zittran, John Palfrey, Urs Gaser, and David Weinberger.

Circular Economy (CE) is a new production and consumption model aiming to redefine processes sustainably by adopting the "take-make-reuse-recycle" concept in all industries, including the architecture-construction industry. Studies show that we spend more than the current amount of Earth's natural resources and need to reduce global resource production and consumption by a third. This paper investigates the "design by research" method to translate theoretical circular economy concepts into practical village-scale design activities in urban + architecture studio education. The studio aims to develop a design model for circular villages that are affordable, sustainable, and socially connected. Circular villages can collectively positively impact the land and each other. For this study, Sille village in Konya– Turkiye has been analyzed from macro-region scale to micro village scale through the year-long M. Arch graduate urban + architecture design studio. Three main approaches were prominent in developing circular/sustainable planning and design strategies for the village: (1) self-sufficiency in terms of resources, food, energy, and workforce, (2) continuity of natural-cultural values and aspects along with the associative collective memory and contextual relationships and (3) utilizing digital tools in the application of circularity concepts. Exploring the potentials and limits of digital tools to accomplish the circularity principles at urban and architectural design levels brings novel insights to the design notion in architectural education.

Cul-de-sacs are examined together with the urban reading in Siverek. Studying cul-de-sacs is instrumental to understand the morphology of Islamic cities. Cul-de-sacs provide a buffer zone between main roads and houses. For this reason, both the privacy phenomenon, which is one of the important issues for Islam, and the safe space need of the residents are important spatial elements. Until 1968, the city of Siverek developed organically within a compact texture of narrow and curvilinear streets, open courtyards, and a adjacent high-walled residences. In this texture, many semi-private cul-de-sacs have an organic form, which is one of the critical spatial elements of the city. Residential walls or courtyard walls form the natural line of traditional streets. In this study, typological analyses of cul-de-sacs were made, and form grammar, a productive method used to analyze architectural language, is included. It aims to examine the relationship between residential settlement and street using the data set created with shape grammars and to make urban propositions for neighbourhood structures in the context of a cul-de-sac using L-systems in the next step. Some parameters have been determined in forming cul-de-sacs that form the urban texture. These parameters were transferred to the model using digital tools. A method thought to be used in urban production has been put forward. The most important reference of this method is dead-end streets.

Figure 1: StyleMe system employs generative adversarial network to automatically generate personalized clothing sketches based on designer's sketches and transfer style from sketch to real clothing image. This figure exhibits the sketch images and clothing images generated by the system. The code, dataset and video demonstration of StyleMe are available at following project link: https://github.com/ExponentiAI/StyleMe.

Data-driven architectural production and operation as explored within Hyperbody rely heavily on system thinking implying that all parts of a system are to be understood in relation to each other. These relations are increasingly established bi-directionally so that data-driven architecture is not only produced (created or designed and fabricated) by digital means but also is incorporating digital, sensing-actuating mechanisms that enable real-time interaction with (natural and artificial) environments and users. Data-driven architectural production and operation exploit, in this context, the generative potential of process-oriented approaches wherein interactions between (human and non-human) agents and their (virtual and physical) environments have emergent properties that enable proliferation of hybrid architectural ecologies.

Data-driven architectural production and operation explored within Hyperbody rely heavily on system thinking implying that all parts of a system are to be understood in relation to each other. These relations are established bi-directionally so that data-driven architecture is not only produced (designed and fabricated) by digital means but also incorporates digital, sensing-actuating mechanisms that ena-ble real-time interaction between (natural or artificial) environments and users. Data-driven architectural production and operation exploit, in this context, the generative potential of process-oriented approach-es wherein interactions between (human and non-human) agents and their (virtual and physical) envi-ronments have emergent properties that enable proliferation of hybrid architectural systems and ecolo-gies.

This paper presents design rationales and prototype of a generative system for visualising possible sequences of information to stimulate human creative thinking. Possible sequences of information segments, usually called contexts, scenarios, narratives, or storytelling, have been used as a tool for exploring and stimulating thinking about possible events, assumptions relating to these events, and courses of actions for a broad range of information designers from public audiences searching on the internet to analysts and policy makers. Thanks to the search technology, it has become easier to obtain information that a person is looking for, however, it is a laborious task to grab an overview of information space so that an information designer can find contextually relevant information pieces and sequence them into contextually meaningful ways. In this research, we design and develop a generative system that visualises possible contexts.

This paper describes research carried out to develop a parametric urban shape grammar for the Zaouiat Lakhdar quarter of the Medina of Marrakech, in Morocco. The goal is to create the basis for a system that could capture some features of the existing urban fabric and apply them in contemporary urban planning

This paper describes research carried out to develop a parametric urban shape grammar for the Zaouiat Lakhdar quarter of the Medina of Marrakech, in Morocco. The goal is to create the basis for a system that could capture some features of the existing urban fabric and apply them in contemporary urban planning

The emergence of networked technologies such as virtual learning environments (VLE's) are providing opportunities for the development of new virtual tools to assist design students, practitioners and researchers in exploring ideas with the aid of visualising and mapping tools and providing interfaces that support interdisciplinary collaboration between design teams. In 1998 a research project was initiated to evaluate the potential of computer assisted learning within Art and Design. This resulted in the development of a virtual learning environment-StudioSpace-designed to support Art and Design students and staff (www.studio-space.net). This paper describes the design process used to develop this VLE and the underlying principles based on a constructivist approach to experiential learning. The ongoing research uses the metaphor of the 'design studio' to explore a range of concepts for generative tools for the representation of design learning, practice and related research. The paper explores some of the ways in which tools such as information retrieval applications, white-boards, visual mapping and digital archives can be combined to provide a virtual online design studio. A further extension to the metaphor provides opportunities for developing new virtual facilities and tools, for example the sketchbook, portfolio, model-making and testing facilities. The virtual design studio has two potential uses: first, to provide a tool box for the design student/practitioner/researcher/educator to undertake collaborative design practice and reflection using CAD/CAM applications; second, to provide systems that help to externalise design methodologies, thus making it possible to gain an insight into the design process itself. Design iteration and the recording of critical decision paths and reflection points can be achieved through the use of meta-data (such as author, date/time created, version number).

Designers immerse themselves in environments rich in inspiration. Previous research has tended to neglect the vital role of sources of inspiration in triggering and guiding designers' activities. This paper reports research which investigates the gathering of inspiration sources and exploration of ideas and hence attempts to understand how inspiration is harnessed. We conducted a progressive series of empirical studies looking at knitwear design: in situ observation; semi-structured interviews; constrained design tasks; and computational modelling. The paper proposes simple general accounts of observed design behaviour and shows how a simple parts-and-relations account can explicate aspects of subtlety and complexity in design.

recurrent problem in many design domains is communication between different members of the design team involved in the different stages of concept creation, embodiment and detail design, fabrication and production. It is frequently the case that misunderstandings arise, that incorrect and/or incomplete specifications are passed from one team group to another, and that errors or inconsistencies in the design are discovered at late stages of the process. Sometimes, design problems are resolved by production or technical staff in ways that are regarded as unsatisfactory by the originating design staff, but further re-corrections may be too late or too costly to attempt. A response to this problem has been to develop broader, more integrated teams, and the introduction of 'concurrent' design processes.

As transformacoes da sociedade proporcionadas pela tecnologia digital nos impelem a estudar quais seriam suas implicacoes no campo da arquitetura. Ha algumas decadas, nota-se o impacto da adocao do desenho auxiliado por computador na apresentacao e representacao de projetos, no entanto, com as evolucoes surgidas com os conceitos de BIM, parametrizacao e sistemas generativos, destaca-se, como campo para investigacao, os possiveis usos destas ferramentas como apoio a tomada de decisoes ao longo do processo de projeto. Neste contexto, esta pesquisa teve como objetivo o desenvolvimento de um algoritmo de projeto com base na legislacao urbana, que possibilite explorar os meios, processos e conhecimentos necessarios para sua criacao, bem como as possiveis implicacoes de sua adocao como suporte ao projeto e simulacao de ocupacoes urbanas. Para o recorte de estudo, foram selecionados os quatro parâmetros definidos pela legislacao que podem ser caracterizados como os principais responsaveis ...

Abstract. The FACADE project is developing a prototype support system for the conceptual design of mechatronic systems, which is designed both to mitigate and exploit the inevitable biasing effects of design tools on design processes. It includes a suite of design environments ...

The design of fashion products can be customised to better meet the needs and preferences of individual consumers. Although the technology to achieve this is becoming available, it is still unclear whether mass customisation of fashion products will be economi-cally viable. This paper explores how the economics of different customisation approaches in the textile industry can be assessed by simulation of the design customisation process. These processes are modelled using the Applied Signposting Model. By simulating with different task durations and under different iter-ation scenarios we show how it is possible to assess the cost of alternative customisation options. This approach to exploring the relationship between prod-uct customisation decisions and the performance of the redesign process has potential to support similar decisions in engineering domains.

Designers immerse themselves in environments rich in inspiration. Previous research has tended to neglect the vital role of sources of inspiration in triggering and guiding designers' activities. This paper reports research which investigates the gathering of inspiration sources and exploration of ideas and hence attempts to understand how inspiration is harnessed.

Producing large numbers of garment variants will only be economically viable if it requires very little human effort. But garment customisation cannot always be fully automated. Applying grading rules maintain the same details but sometimes achieves a different overall effect; but the customer expects the same overall effect and is less concerned about details. Choosing between alternative customisations requires a human designer's trained perceptual judgement. Therefore a viable mass customisation support system must support the repeated redesign of a garment by combining automatic design with fast human editing. Evaluating and modifying the suggestions of others is a natural and efficient activity for designers. This paper describes two prototype automatic design systems exploring techniques that could be used for mass customisation of knitted garments – in which the shape and patterns are indivisibly linked. An early pattern placing system that automatically altered both shap...

As commercial knitwear designers gain experience, many appear to lose their creativity. They often find they have gone stale, and need to change jobs to find fresh challenges. But although designer burnout is a significant phenomenon, experienced designers gain both a broader understanding of their design context and thus a more sophisticated understanding of design problems, and develop expertise in creating the designs their companies require quickly and efficiently. However they may find it harder to create innovative designs. Partly as a result of widespread fallacious beliefs about creativity being antithetical to rational problem solving, experienced designers are not encouraged to develop sufficiently flexible skills, and the skills they do develop are undervalued. This paper relates the demands that knitwear designers face to the cognitive psychology of learning and expertise, to examine what designers learn from experience, and how the demands that govern their designing be...

The meaning and importance assigned to women’s visual privacy is changing day by day and women try to protect their visual privacy in these days. Women’s visual privacy is connected to their house design, religion (in Islamic religion women is visual privacy is more important than other in religions), and socio-economic classes....etc. In addition to these, architectural elements affect women’s visual privacy also. The author’s main concern in this research is to analyze women’s visual privacy according to different points of view in two different streets of the city which includes traditional and modern apartment’s neighboorhoods. Abdullah Pasa Street (where traditional houses are located) and Yildiz Street (where modern apartments are located) have been taken into consideration as sample streets in Famagusta (North Cypus). A detailed questionnaire was done with the users of these buildings. As a result, new suggestions are proposed for improving visual privacy.

This paper presents our on-going project called "Generative Website Project". We design and develop generative systems that visualise possible contexts, that is, possible sequences of information out of existing information in order to stimulate human creative thinking. Possible sequences of information segments, usually called storytelling, scenarios, narratives, or, have been used as a tool for exploring and stimulating thinking about possible events, assumptions relating to these events, and courses of actions for a broad range of information designers from public audiences searching on the internet to analysts and policy makers. In this paper, we report our progress of the Generative Website Project and related technology.

In [5] a new approach to intelligent data analysis [1] based on information flow distribution study in a flow graph was presented and studied. Branches of the flow graph are interpreted as decision rules, whereas the flow graph is supposed to describe a decision algorithm. We propose ...

This paper presents our ongoing project called "Generative Website Project". We design and develop generative systems that visualise possible contexts, that is, possible sequences of information out of existing information in order to stimulate human creative thinking. Possible sequences of information segments, usually called storytelling, scenarios, narratives, or, have been used as a tool for exploring and stimulating thinking about possible events, assumptions relating to these events, and courses of actions for a broad range of information designers from public audiences searching on the internet to analysts and policy makers. In this paper, we report our progress of the Generative Website Project and related technology.

This paper presents a method for visualising possible sequences of information. Possible sequences of information segments, called contexts, scenarios, narratives, or storytellings, have been used as a tool for exploring and stimulating thinking about possible events, assumptions relating to these events, and courses of actions, i.e. chances for a broad range of information designers from public audiences searching on the internet to analysts and policy makers. Thanks to the search technology, it has become easier to obtain information that a person is looking for, however, it is a laborious task to grab an overview of information space so that an information designer can find contextually relevant information pieces and sequence them into contextually meaningful ways. In this research, we design and develop a series of generative systems that visualises possible contexts.

Sources of inspiration, such as garments, works of art or natural objects, are used throughout the knitwear design process to establish the context for a new season, to provide a basis for concrete design and to set colour schemes. Through a combination of observational studies, experimental techniques and computer modelling, the 'Mechanisms of Inspiration in Novel Design' (MIND) project seeks to understand the use of sources of inspiration in the knitwear industry. The goal of the research is to provide a basis for improved support for designers in terms of computer tools, organisational support and work-oriented education.

Drawing ideas from previous designs and other sources of inspiration is a universal part of all human designing, but the forms that adaptation takes, and the ways it is conceptualised both differ between industries. This paper describes how sources of inspiration are used in commercial knitwear design. Knitwear designers actively search for sources of inspiration, which they use both to define the space of designs that will meet their purposes and tune their tacit perceptions of what is appropriate, and in the development of individual designs. They employ active strategies for guiding their idea generation actions to produce the ideas the situation requires. Knitwear designers routinely use a very broad range of sources of inspiration, and are explicitly aware and open about how they use them. Identifying a good source is often the key creative step in knitwear design, and being able to do this well is a vital skill for a designer. The paper makes some provisional generalisations about the structure of design processes driven by thinking with sources of inspiration, to provoke and facilitate comparisons with other industries.

T he Enterprise-D encounters a starship crewed by an alien race with whom Captain Picard and company cannot communicate despite the omnipotence of their universal translator 1. Eventually they discover that this race communicates entirely by reference to historical anecdotes or shared myths. The Tamarians can express in a few words a complex concept like 'two strangers overcome their differences in the face of great peril and are friends ever after', while conjuring up a rich picture of the emotions involved in the adventure. However it takes them equally long to convey a simple concept such as 'give me a knife', by referring to a story of a man who opened his arms in front of a town wall and was given something. In an ethnographic study of the knitwear design process we have observed an almost equal reliance on shared cultural references to express design 1 Star Trek: The Next Generation Series 5 Episode 2 'Darmok', first broadcast in the week of

Intellectual health like physical health depends on a balance between feeding and exercise. But the demands of work often make a healthy balance very hard to maintain. In a detailed study of the knitwear industry and in comparisons with other design fields, we have observed that designers use external sources of ideas for a variety of different purposes throughout the design process. We have also observed that designers' creative potential is often limited by the failure of managers to understand design, and especially their designers' need to search for sources of new ideas. Managing design processes to maximize creative output requires both an understanding of the role of sources of ideas in the creative process, and proactive management to support and encourage the renewal of intellectual resources. In this paper we suggest some practical steps for supporting continuing creativity.

This paper argues that the widespread belief that ambiguity is beneficial in design communication stems from conceptual confusion. Communicating imprecise, uncertain and provisional ideas is a vital part of design teamwork, but what is uncertain and provisional needs to be expressed as clearly as possible. Understanding what uncertainty information designers can and should communicate, and how, is an urgent task for research. Viewing design communication as conveying permitted spaces for further designing is a useful rationalisation for understanding what designers need from their notations and computer tools, to achieve clear communication of uncertain ideas. The paper presents a typology of ways that designs can be uncertain. It discusses how sketches and other representations of designs can be both intrinsically ambiguous, and ambiguous or misleading by failing to convey information about uncertainty and provisionality, with reference to knitwear design, where communication using inadequate representations causes severe problems. It concludes that systematic use of meta-notations for conveying provisionality and uncertainty can reduce these problems. KEYWORDS Collaborative design, design communication, sketching, meta-notation, knowledge level, knitwear 'Ambiguous' is an ambiguous word: its standard meaning is 'interpretable in two or more distinct ways'; but it is also used to mean 'vague or imprecise'. The ambiguity of 'ambiguity' causes confusion. Just as the communication of design ideas in real life suffers from a failure to understand the nature of the problem (at least in commercial knitwear design, the industry we have studied in detail), academic analyses of 'ambiguity' in design have been muddied by the conflation of different types of imprecision into an over-broad concept of 'ambiguity'. This makes some published discussions of ambiguity in design communication impossible to interpret, even when they stem from

Communication between different members of a design team often poses difficulties. This paper reports on the results of a detailed empirical study of communication in over twenty British, German and Italian knitwear companies. The knitwear design process is shared by the designers, who plan the visual and tactile appearance of the garments, and the technicians, who have to realise the garment on a knitting machine. They comprise a typical but small design team whose members have different backgrounds and expertise. Knitwear design allows a detailed analysis of the causes and effects of communication breakdown. Designers specify their designs inaccurately, incompletely and inconsistently; technicians interpret these specifications according to their previous experience of similar designs, and produce garments very different from the designers' original intentions. Knitwear is inherently difficult to describe, as no simple and complete notation exists; and the relationship between visual appearance and structure and technical properties of knitted fabric is subtle and complex. Designers and technicians have different cognitive approaches and are very different people. At the same time the interaction between designers and technicians is badly managed in many companies. This paper argues that improving the accuracy and reliability of designers' specifications would significantly enhance the design process. It concludes with a description of the architecture of an intelligent automatic design system that generates technically correct designs from the designers' customary notations.

Sources of inspiration, such as garments, works of art or natural objects, are used throughout the knitwear design process to establish the context for a new season, to provide a basis for concrete design and to set colour schemes. Through a combination of observational studies, experimental techniques and computer modelling, the 'Mechanisms of Inspiration in Novel Design' (MIND) project seeks to understand the use of sources of inspiration in the knitwear industry. The goal of the research is to provide a basis for improved support for designers in terms of computer tools, organisational support and work-oriented education.

Design process models are generated for a variety of purposes. Most design process models are abstract and only include large-scale tasks. This paper reports on the experience of building a generic hierarchical flow chart model of design processes in knitwear design based on a study of processes in 25 companies with a wide spectrum of market positions. The model shows tasks, decision point and iteration loops. It does not show differences in task duration and effort, or the frequency or depth of iteration, which account for the differences in the observed processes. Variations arise from the effort that individual companies invested in the design process and their willingness and ability to allow iteration. Detailed properties of the product also influenced iteration and the distribution of effort. This research shows that considerable modelling effort is required to capture characteristics of specific processes, but that generic models can be used as a starting point.

Designers often explain new concepts and new ideas by reference to existing designs. This is parsimonious, as it only requires a pointer to the referent and a description of the modifications. Such descriptions can be extremely powerful, expressing the entire context of a design or a process in a few words. However similarity assertions are inherently ambiguous, because they depend not only on the chosen description but also on the intention behind the similarity comparison. In this paper we attempt to analyse the effect that the ambiguity of similarity references has on communication and idea generation in design. The reinterpretation of a similarity assertion can be extremely creative, where ambiguity allows for new interpretations of a problem. At the same time, it can make accurate communication extremely difficult because every assertion can be interpreted differently unless the context is fully shared.

The design of knitted garments is an activity shared by knitwear designers (who are almost all young and female) and knitting machine technicians (who are almost all male and usually older). The process involves programming knitting machines using CAD systems, which are designed for and used by the technicians. The designers get much less training and access to the CAD systems than they want. This paper examines why this is, and what would be involved in creating a situation where the designers are empowered over the technology of knitwear design. It concludes that the limitations of the technology cause this situation, which is reinforced by economic and attitudinal factors.