Distinction Award in the MA Design Practice

Northumbria University awarded me Distinction in its MA Design Practice for the thesis Creating Innovative Design Software Solutions within Collaborative/Distributed Design Environments (click here to download the thesis):

The current status of the software industry – where most products are globalized and the development is distributed through teams around the world, across different continents and time zones – poses a problem to interaction designers: knowledge and expertise of all parties involved must be exploited, no matter how these parties are distributed geographically and organizationally.

In this project, we’ve analyzed the current interaction design processes and methodologies at Autodesk® by reviewing the development cycle of one product, contrasting it with current research findings in the field of collaborative/distributed design, identity areas for improvement and propose strategies for supporting creativity and innovation in this process.

A number of interventions were proposed, and some considerations were made based on a hindsight review of strategies implemented to improve processes.

Design Practice and Management

As an User Experience Design Manager for Autodesk® – world’s leading design software and digital content, and proud producer of successful products like AutoCAD®, 3Ds Max® and Maya® – I was interested in understanding how to design interfaces of software that would help our customers be more creative and foster innovation.

The challenges are many: first, given the nature of the software we design, a better understanding of the nature of creativity and the processes that underpin it are needed in order to be able to make contributions to the design of creativity support tools (Johnson & Carruthers, 2006).

Second, given the current status of the software industry – where most products are globalized and the development is distributed through teams around the world, across different continents and time zones – and the networked computing interaction paradigm with which our designers have to work on, we must learn to exploit the knowledge and expertise of all parties in the design team, no matter how distributed geographically and organizationally these parties might be (Lang, Dickinson, & Buchal, 2002).

Collaborative vs Distributed Design

Globalization of development processes is based on the principal of making the most efficient use of resources possible for whatever task needs to be done. In product design, this principal means exploiting the knowledge and expertise of all parties involved, including marketing, engineering, design, management, suppliers, production, etc., in the design team, no matter how these parties are distributed geographically and organizationally (Lang, Dickinson, & Buchal, 2002). The same logic applies to software development

In response to this increasing need to assist collective work in an information technology context, recent studies have shifted their foci toward cooperative work, concerning to the creation of new technical-organizational systems, which support collective work, greater interaction between design stakeholders, as well as capitalization and reuse of design knowledge (Détienne, 2006).

Falzon (Falzon, Montmollin, & Béguin, 1996) has stressed out a distinction between two design situations according to the nature of shared goals: co-design (or collaborative design) and distributed design.

Collaborative Design

In Collaborative design, design partners develop the solution together – they share an identical goal and contribute to reach it through their specific competences, they do this with very strong constraints of direct cooperation in order to guarantee the success of the problem resolution. The competence of the partners can vary depending on the level of competence (e.g. interaction between designers of different seniority) or on the type of competence (e.g. interaction between drafters and engineers). Design solutions are not only based on purely technical problem solving criteria. They also result from compromises between designers: solutions are negotiated.

Distributed Design

In Distributed Design, the actors of the design who are simultaneously (but not together) involved on the same cooperation process carry out well determined tasks. Such tasks having been allocated beforehand, and they pursue goals (or at least sub-goals) that are specific to them and have as an objective to participate as efficiently as possible in the collective resolution of the problem. Distributed design is typical for concurrent engineering in which the various sides of the production system must function in strong synergy during the product development cycle.

Problem Framing

Information Technologies supporting Creativity

Shneiderman (2000) in developing his Genex framework advocates that software designers can create support tools that enable users to be more creative.

The Genex (generator of excellence) proposal consists of four phases:

• Collect: learn from previous works stored in libraries, the Web, etc.
• Relate: consult with peers and mentors at early, middle, and late stages;
• Create: explore, compose, evaluate possible solutions; and;
• Donate: disseminate the results and contribute to the libraries.

These phases also support the cyclical nature of the process in that dissemination of results should support future users who seek to learn from previous work. Shneiderman (2000) proposes that with careful design it is possible to combine inspirationalists’, structuralists’ and situationalists’ perspectives.

Within this framework, Shneiderman proposes eight activities software tools should support:

• Searching and browsing digital libraries;
• Visualizing data and processes;
• Consulting with Peers/mentors;
• Thinking by free association;
• Exploring solutions (“what-if” tools);
• Composing artifacts and performances;
• Reviewing versions;
• Disseminating Results.

Distributed Design Issues

Since the distribution of tasks has been identified as means of minimizing software development risks, let’s analyze the current interaction design process at through Distributed Design perspective. Lang, Dickinson and Buchal’s Cognitive Factors in Distributed Design seem an appropriate framework to start with (2002), which divides the design process through distributed teams into 5 areas, namely design methodology, collaboration, teamwork, knowledge management and design representation.

Combining Frameworks

Merging the both previously mentioned frameworks (Shneiderman’s Genex, and Lang, Dickinson and Buchal’s Cognitive Factors in Distributed Design) allowed me to capture enough aspects of the design process in distributed teams, so that I could start analyzing the design practice at Autodesk®.

Identifying Issues and Stakeholders

I needed to find a way to – by looking at the different aspects of the design practice at Autodesk – identify and rank the issues (be them enablers or barriers), as well as their corresponding stakeholders. So I decided to look to at the design practice at Autodesk for the organization strategy’s point of view.

An organization’s strategy may be examined using multiple criteria (Mintzberg, 2003). Focusing multiple lenses on a given phenomenon – in my case here, different aspects of a design process – highlights various elements of that phenomenon (Ancona, 1998). Each lens suggests a different set of practices and solutions to managers.

The lenses I’ve chosen to analyze the organization strategy were – and a few assumptions I had about them – were:

Social/Cultural

Those issues related to the cultural and social gap between Western designers and Chinese designers background. As Product Design Manager, I could facilitate their collaboration and be a “bridge”: that could provide me with insights of how can I help they improve their collaboration with their overseas peers;

Political/Economical

Identifying key players who could influence practices/policies (like training, processes, decision making rights, etc…) would give me insights of who are the stakeholders I’d need to get buy in for improvement recommendations;

Organizational

Those issues related to the current practices/policies in place at Autodesk, and how these foster/hinder collaboration between my team, their overseas peers, as well as with others teams; that could give me insights about what processes need to be improved;

Technical

Those issues related to technological hindrances for collaboration of design distributed teams; looking at these issues could give insights of which tools/training needs to be provided/better used by the team in order to improve collaboration.

Ranking of Issues

In order to analyze the design practice at Autodesk from all these different lenses simultaneously, I’ve decided to use a card sorting technique (Spencer, 2004). I’ve invited my peer Douglas Wang – also a User Experience Design Manager for Autodesk in Shanghai – to help me rank each issue according to the corresponding design practice at Autodesk in a scale from 1 to 5 – where 1 was “low” and 5 “high” – into (S) social/cultural, (P) political/economical, and (O) organizational, and (T) technical.

My assumptions were that issues that ranked – for example – high on the Social (s) would related to my team’s culture and, therefore could be dealt locally); Political/economical (P) would involve higher level of management; Organizational (O) are related to the processes at Autodesk; Technological (T) would relate to the skills/facilities/training (or lack of!) that support the design.

Contextual Relationships Maps

In order to visualize the relationship between the issues, I decided to produce Contextual Relationship Maps. Given the massive amount of data produced and the complexity of the relationships (several issues were recurrent in different aspects of the design process), I’ve broken down the process of producing the maps into steps:

• Making individual maps for each lens (social, political, organizational, technical);
• Comparing/contrasting the maps;

Four Contextual Relationship Maps were produced — Social, Political, Organizational, and Technical — by listing the highest ranking issue on each line of the spreadsheet (check the appendix B: contextual relationship maps).

Each of the most “connected” issues were classified either as related to:

• Tools;
• People;
• Processes;
• Artifacts.

These four maps allowed to dissect the design practice at Autodesk its very essential aspects, as well as to point the different stakeholders in the organization.

Social

Focusing only on the SOCIAL/CULTURAL ISSUES, I’ve noticed that communication seem to play in important role on the practice; not just as language issues, but all the other aspects of communication that facilitates members of different teams to stay on the same page. Some of these concerns include having shared understanding of design problem spaces, knowing how to better convey design rationale/intent during feedback sessions/design reviews, etc…

Based on the insights highlighted on the map, my assumptions are that we should look into ways of facilitating Cognitive Synchronization through better support to multiple means of communications, such as Emails and Instant Messaging, and expanding their Problem Analysis skills and Thinking Tools, which enables them and their partners to reach two objectives:

• Assure that they each have knowledge of the facts relating to the state of the situation/problem data, state of the solutions, accepted hypothesis, etc;
• Assure that they share a common knowledge regarding the domain – technical rules, objects in the domain and their features, resolution procedures, etc.

Political

Focusing only on the POLITICAL/ECONOMICAL ISSUES, I’ve noticed that knowledge management issues related to supporting decision making seem to play a strong role

In the context of distributed design – in which design teams and other functional groups have to collaborate with Product Management to turn Marketing Requirements into Product Requirement Documents – it’s only natural that Product definition occurs at the beginning of the project, which is precisely when the team has the least experience in working together, making dispersed meetings less effective (Smith & Blanck, 2002).

Therefore, Designers need to facilitate decision making by assuring stakeholders of the quality of the artifacts produced and the innovative aspects of their design solutions (hence, the Collaboration with Product Managers and the Qualitative User/Customer Concept Validation Methods).

Organizational

Focusing only on the ORGANIZATIONAL ISSUES, I’ve noticed that collaboration processes that facilitate the production of artifacts seem to play an important role:

Given the distributed nature of the design process at Autodesk, the efficiency of producing artifacts that clearly communicate design intent and rationale is crucial to inform the decision makers (check Political Map);

This is where the Social Issues start affecting the Organization: producing quality design artifacts that clearly communicate design intent and rationale will be tightly coupled with the ability of junior designers (as individual contributors, or Authors of the design artifacts) to Modularize / Decompose design problems, of their skills of Argumentation / Justification during Presentations they do for Design Reviews, which are fundamental aspects of the problem solving framework of the design practice.

Technical

Focusing only on the TECHNICAL ISSUES, I’ve noticed that both tools for coordinating processes, as well as representation methods for communicating design history and rationale seem to play a strong role:

In the “traditional” co-located design process, designers have multiple ways of representing there design ideas: sketching, storytelling, prototyping.

Management Interventions

A number of interventions were proposed, and some considerations were made based on a hindsight review of strategies implemented to improve processes in each area:

Social

The problems that designers face are rarely straightforward and clear. They are complex and lack “right answers.” Skillful professional practice often depends less on factual knowledge than on the ability to reflect before taking action (Schön, 1992).

I believe that, by introducing smaller design deliverables more closely spaced in between the milestones and easily shared with other designers/team, we could address the challenges mentioned above:

• At the individual level, designers will be able to better monitor their own progress of their understanding of the design problem;
• At the design team level, designers will rapidly iterate their designs with other members on the team, who – in turn – will be able to understand better and give more insightful feedback to smaller chunks of design, facilitating cognitive synchronization;
• At the management level, managers will make better use of their bandwidth by better assessing the individual needs for each designer, and provide the appropriate coaching;

Political

In regards to ensuring that junior designers at Autodesk are using the appropriate methodology, one of the challenges that we face – to exploit the knowledge and expertise of all parties in the design team, no matter how distributed geographically and organizationally these parties might be (Lang, Dickinson, & Buchal, 2002) – also works as an opportunity. Junior designers in China have access to a pool of over 80 people in the global User Experience Design team, from which they can learn from when it comes to design methods.

With regards to the challenge of improving the flow of information, I believe that there are opportunities to benefit from Web 2.0 related tools previously mentioned (SharePoint® sites, wikis, Intranets, etc…) by pursuing a “push-approach” to information gathering, such making use of RSS feeds, Alerts, and/or Feed Aggregators.

Organizational

With regards to ensuring that junior designers at Autodesk improve the way they analyze design problems, I believe there are plenty of opportunities (in alignment with Political Issues previously mentioned) for Junior designers in China to leverage the expertise of other designers across the company. They have access to a pool of over 80 people in the global User Experience Design team, from which they can learn when it comes to design methods.

Regarding Presentations done during Design Reviews, there are opportunities for complementing the Design Journals (strategy suggested to address some of the Cultural Issues) with other reflective practice templates, like the use of checklists, etc..

Technical

In regards to ensure that designers keep using sketches for improved evaluation / re-stating of design problems even in a distributed environments, I believe that are opportunities for us to leverage on communication technology already at our disposal at Autodesk – like the annotation features of LiveMeeting™ –or explore the possibility of acquiring full-fledged collaborative design systems, like Interactive WhiteBoards.