title: "The Essence of Program Semantics Visualizers: A Three-Axis Model"
authors: Josh Pollock, Grace Oh, Eunice Jun, Philip J. Guo, Zachary Tatlock
venue: Workshop on Evaluation and Usability of Programming Languages and Tools (PLATEAU)
year: 2020

abstract: >
  A program semantics visualizer (PSV) helps illuminate a language's
  semantics by explaining the runtime execution of programs. PSVs are
  often used in introductory programming (CS1) courses to help introduce
  a notional machine, an abstraction of the computer that executes the
  language. But what information should PSVs present to fully explain
  such notional machines? In this paper we propose a three-axis model to
  assess the design of PSVs that visualize execution traces. PSVs should
  help users by clearly answering three questions: What is the machine's
  configuration at each execution step? Why did an execution step take
  place? How did an execution step change the machine's configuration?
  We demonstrate our model's utility for assessing PSVs by explaining
  why, in actual classroom use, instructors have resorted to manually
  extending PSV output. In particular, we study instructors' additions
  to visualizations generated by Python Tutor, the most popular PSV.

bibtex: >
  @inproceedings{Pollock2020,
    title={The Essence of Program Semantics Visualizers: A Three-Axis Model},
    author={Pollock, Josh and Oh, Grace and Jun, Eunice and Guo, Philip J. and Tatlock, Zachary},
    booktitle = {11th annual workshop on the intersection of HCI and PL},
    series = {PLATEAU '20},
    year={2020}
  }