Adrian Kuhn

Adrian Kuhn, Robert DeLine

Modern software development requires a large investment in learning application programming interfaces (APIs). Recent research found that the learning materials themselves are often inadequate: developers struggle to find answers beyond simple usage scenarios. Solving these problems requires a large investment in tool and search engine development. To understand where further investment would be most useful, we ran a study with 19 professional developers to understand what a solution might look like, free of technical constraints. In this paper, we report on design implications of tools for API learning, grounded in the reality of the professional developers themselves. The reoccurring themes in the participants' feedback were trustworthiness, confidentiality, information overload and the need for code examples as first-class documentation artifacts.

Adrian Kuhn, Gail C. Murphy

In a recent empirical study we found that evaluating abstractions of Model-Driven Engineering (MDE) is not as straight forward as it might seem. In this paper, we report on the challenges that we as researchers faced when we conducted the aforementioned field study. In our study we found that modeling happens within a complex ecosystem of different people working in different roles. An empirical evaluation should thus mind the ecosystem, that is, focus on both technical and human factors. In the following, we present and discuss five lessons learnt from our recent work.

Adrian Kuhn, Peter Loretan, Oscar Nierstrasz

Software visualizations can provide a concise overview of a complex software system. Unfortunately, since software has no physical shape, there is no "natural" mapping of software to a two-dimensional space. As a consequence most visualizations tend to use a layout in which position and distance have no meaning, and consequently layout typical diverges from one visualization to another. We propose a consistent layout for software maps in which the position of a software artifact reflects its \emph{vocabulary}, and distance corresponds to similarity of vocabulary. We use Latent Semantic Indexing (LSI) to map software artifacts to a vector space, and then use Multidimensional Scaling (MDS) to map this vector space down to two dimensions. The resulting consistent layout allows us to develop a variety of thematic software maps that express very different aspects of software while making it easy to compare them. The approach is especially suitable for comparing views of evolving software, since the vocabulary of software artifacts tends to be stable over time.

Adrian Kuhn, Gail C. Murphy, C. Albert Thompson

In this paper, we investigate model-driven engineering, reporting on an exploratory case-study conducted at a large automotive company. The study consisted of interviews with 20 engineers and managers working in different roles. We found that, in the context of a large organization, contextual forces dominate the cognitive issues of using model-driven technology. The four forces we identified that are likely independent of the particular abstractions chosen as the basis of software development are the need for diffing in software product lines, the needs for problem-specific languages and types, the need for live modeling in exploratory activities, and the need for point-to-point traceability between artifacts. We also identified triggers of accidental complexity, which we refer to as points of friction introduced by languages and tools. Examples of the friction points identified are insufficient support for model diffing, point-to-point traceability, and model changes at runtime.