Engineering design thinking combines concepts from heterogeneous sources like personal experience, colleagues, digital and hardcopy media. Despite this challenge, modes of thinking across levels of abstraction through multi-dimensional (spatial) representations are widely neglected in digital support systems. This paper aims to summarize lessons learned through years of experience with software tools that augment this visio-spatial conceptual thinking. This work cuts across disciplines to provide a needed, coherent starting point for other researchers to examine complex outstanding issues on a class of promising support tools which have yet to gain widespread popularity. Three studies are used to provide specific examples across design phases, from conceptual design to embodiment. Each study also focuses on an exemplar of diagrammatic software: the University of Cambridge Design Rationale editor (DRed), the Institute for Human Machine Cognition's (IHMC) CmapTools and the Open University's Compendium hypermedia tool. This synthesis reiterates how hypermedia diagrams provide many unique, valuable functions while indicating important practical boundaries and limitations. Future research proposed includes: a need to build more diagrammatic literacy into engineering practice, the need for more detailed studies with experts in industry and specific directions for refining the hypermedia diagram software interfaces.

Engineers use many "cognitive technologies" to bridge the gap between human cognitive limits and the requirements of complex, distributed, asynchronous and information-intense cooperative work. This paper examines the value of narrative in engineering work to improve the design of emerging diagram-based software tools for sensemaking and information navigation. These have been studied and even adopted by some aerospace organizations but they are not yet common. In some cases, these aim to replace more traditional, narrative documents, like textual reports, because graphs reflect better the hyperlinked, non-linear structures in both engineered systems and information repositories. It is critical, however, to determine what is lost and what is gained by that transition. This paper presents a case study of teams at an aerospace company as they begin regular use of Design Rationale editor (DRed) diagramming software. The engineers were observed and interviewed over several weeks. Results, structured and analysed using DRed and the cognitive dimensions framework, suggested that diagrams are a better way to build and organize work incrementally while narrative structure is valuable in verifying completeness of work and initially guiding users around documents. Further work will involve more detailed study and modification of DRed to exploit the value of narrative.

Graph representations are emerging to replace more traditional narrative forms of knowledge capture. These seem to be a better fit for work with modern hyperlinked, non-linear computerized systems and design problems. The old forms, however, possessed important characteristics for effective knowledge capture. It is not exactly clear what is lost and gained in the transition. This paper explores narrative theories and examines them using the cognitive dimensions framework. A review of literature finds that deliberately constructed narrative structure is key to affording meaningful experience of representations. This suggests flaws in current documentation practice. Proposed concepts situate many new questions that will require further study.

Longer lifecycles and a shifting industrial focus from simple product delivery to through-life support have increased the need to retain organisational knowledge for the duration of product lifecycles and beyond. This fragmentation of work across time and organizations buries organizational understanding of design processes and of the underpinning design rationale in many disparate representations. Further, existing report-based documentary practices tend to omit key information which could compromise an engineer's ability to comprehend and reuse this information in addition to requiring significant additional authoring work. Emerging approaches to documentation are discussed and compared as cognitive technologies in a snowmobile drive shaft example. One design episode is assisted by interactively documenting decisions in an IBIS-based tool whose output forms the record. Another is passively documented via an activity-modelling approach and media-enhanced records. This work provides examples of situating discussions about engineering design documentation practices and describes improvements for further development and testing of both the activity-modelling and media-enhanced records and the IBIS-based approaches.