CS 609 Diskussionsseminar: Human Factors in Programming Languages

Blackwell, First Steps in Programming: A Rationale for Attention Investment Models, HCC 2002.

This paper serves two roles. First, it provides definitions of “programmer” and “programming” that are somewhat broader than you might expect. This will help us to define the scope of the seminar. In particular, we will consider many aspects of what might traditionally be called “user interface design” to be within our purview as language designers. Second, it introduces the attention investment model, which can help to understand when a user will invest effort in learning a new language or feature, and suggests strategies for encouraging this investment.

Green and Petre, Usability Analysis of Visual Programming Environments: A ‘Cognitive Dimensions’ Framework, JVLC 1996.

The cognitive dimensions are the most well-known and widely used analytical framework for the usability of languages and interfaces. Perhaps their most valuable contribution to the field is providing a shared vocabulary for notational qualities that impact usability, something that will also be useful for us in this seminar. Besides this vocabulary, the framework emphasizes the trade-offs between dimensions.

Peyton Jones, Blackwell, and Burnett, A User-Centered Approach to Functions in Excel, ICFP 2003.

This paper applies the two intellectual tools we have seen so far–the Attention Investment model and the Cognitive Dimensions–to the design of language extensions to Excel spreadsheets. It provides a nice illustration of how these tools can support a systematic, analytical approach to user-focused design.

Wilson et al., Harnessing Curiosity to Increase Correctness in End-User Programming, CHI 2003.

This paper introduces the surprise-explain-reward strategy in the context of encouraging end-user programmers to test their programs. Attention investment provides a theory about how users decide where to spend their attention based on cost, risk, and reward. Surprise-explain-reward provides a strategy for altering this equation. Specifically, it attempts to lower the (perceived and actual) costs associated with learning a new feature, while making the reward more immediate and clear.

Moody, The “Physics” of Notations: Towards a Scientific Basis for Constructing Visual Notations in Software Engineering, TSE 2009.

The physics of notations is a theory for designing usable visual languages. It is founded on a synthesis and organization of a large amount of existing research on human perception and cognition. Compared to the cognitive dimensions, it is intended to be prescriptive, more objective, and falsifiable. However, it focuses mostly on picking effective graphical symbols, and so is less generally applicable than the cognitive dimensions and the other theories we have discussed.

Burnett et al., Gender Differences and Programming Environments: Across Programming Populations, ESEM 2010.

The field of Gender HCI is concerned with how males and females interact differently with computers (on average), and how to create software that takes these differences into account, thereby supporting a broader range of users and problem-solving strategies. I think this is an especially important issue for programming languages and tools since these are the keys to computing empowerment. This paper looks specifically at the intersection of Gender HCI and programming environments. The paper is also highly empirical, complementing the mostly theoretical papers we have read so far in this seminar.

Pane and Myers, More Natural Programming Languages and Environments, in End User Development, Springer, 2006.

A language can be called “natural” if it has a high closeness of mapping with the (possibly informal) way that (some group of) people think about and express solutions to programming-like problems. The term can be a red flag in human-centric research, since it is usually applied post hoc and without evidence. The Natural Programming Project, however, proposes a design methodology where designers first identify what is “natural” for users in a particular domain, then iteratively design and evaluate a language tailored to the way these potential users already think. While their focus is on non-programmers, the methodology can also be applied in other contexts. In contrast to the other design tools we have discussed this semester, the natural programming methodology is fundamentally user-centric and bottom-up.

Ko and Myers, Designing the Whyline: A Debugging Interface for Asking Questions about Program Behavior, CHI 2004.

As part of the Natural Programming Project, Ko and Myers found that, when faced with unexpected results, both programmers and non-programmers formulate “why did …” and “why didn’t …” questions about the program’s behavior. With traditional debuggers, attempting to answer these questions is difficult and time-consuming, perhaps because the interaction with the system is in quite different terms (e.g. breakpoints, log files, code-stepping). The Whyline allows users to ask these kinds of questions directly through a process called interrogative debugging. This paper also illustrates how the cognitive dimensions can be used in interaction design, in addition to interface and language design.

Coughlan and Johnson, Interaction in Creative Tasks: Ideation, Representation and Evaluation in Composition, CHI 2006.

Programming is largely about creative problem solving, but promoting creativity is usually not an explicit design goal of programming languages and tools. In the context of music composition, this paper presents a model of human creativity and describes how tools can be designed to support this process. It focuses especially on supporting collaborative creativity.

Meyerovich and Rabkin, Empirical Analysis of Programming Language Adoption, OOPSLA 2013.

What makes a programming language popular? I think programming language academics are painfully aware that the most popular languages lack many of the properties and qualities we care about. This paper presents a ton of data about how and why languages become popular. It focuses especially on the role of social factors and existing infrastructure in language success, but also looks at the public perception of various language features. Understanding the programming language market is an important part of getting our ideas used by wider world.