Lessons Learned from Using Reprogrammable Prototypes with End-User Developers

Abstract

Involving end-users in the development of a product before it is deployed has great potential to increase the fit between a product and individual users' needs. While end-users can be directly involved in modifying low-fidelity prototypes, they are left out when it comes to high-fidelity interactive prototypes -in part because these cannot be modified directly or require time-consuming edit-compile-run cycles. High-fidelity prototypes, however, are more engaging for users. We created a reprogrammable high-fidelity prototype and explored its use in shortterm prototyping workshops with end-user developers, i.e. end-users with programming experience, in the domain of collaborative writing. We report observations and pitfalls, and distill four lessons learned into guidelines on how to use reprogrammable high-fidelity prototypes with end-users in contexts with limited resources. Our experiences demonstrate, among other things, that reprogrammable high-fidelity prototypes are difficult to work with -even for experienced programmers -and emphasize the need for careful attention to guiding participants, time for familiarization, and catering to multiple levels of programming experience.

Between Principle and Pragmatism: Reflections on Prototyping Computational Media with Webstrates

“…This blurs the distinction between using and developing an application as both happens in the same environment. However, the literate computing aspect may cause confusion and prevent people from using their prior programming knowledge, as it differs strongly from conventional code editors or IDEs (Integrated Development Environments) [6].…”

“…This avoided the need to switch between different views and narrowed the distance between development and use. Literate computing, however, is an atypical model for developing applications and therefore it could be sometimes difficult to mentally separate what is part of the development environment and what is part of the application, e.g., confusing parts of the UI of Codestrates v1 with the UI of a word processor implemented in it [6].…”

Webstrates, Codestrates v2, and Varv: A Software Stack for Computational Media

“…With the Codestrates v2 platform it is, e.g., trivial to implement a computational notebook on top of Webstrates. 3 Varv 4 [4], finally, adds a programming model for computational media, which allows for inherently and incrementally extensible software. Varv represents reprogrammable interactive software as a declarative data structure.…”

“…Throughout the years, we and others have used Webstrates, Codestrates v1, and Codestrates v2 to explore computational media in a variety of projects including collaborative data visualization [1,8,12], public libraries [6,15], collaborative video editing [10], collaborative programming assignments [5], computational laboratory notebooks [13], collaborative writing tools [3], affinity diagramming [11], and hybrid meetings [7].…”

Casting the runes and parsing them: Unpacking software mediation, interactions, and computational literacy in non-conventional programming configurations

“…And while there are other established terms such as end-user developers (Fischer and Giaccardi 2006;Borowski and Larsen-Ledet 2021), this discourse implicitly creates and sustains a particular relationship between people and artifacts. For instance, in the case of computational media, the output can be other computational media.…”