探索设计空间:用于探索式编程的并行版本
Probing the Design Space: Parallel Versions for Exploratory Programming
AI总结:
本文提出 Exploriants,一种扩展基于示例的实时编程的工具,通过变点和比较视图帮助探索式程序员结构化地评估进展,并在图像处理、数据处理和游戏开发中验证其有效性。
摘要:
探索式编程涉及开放式任务。为了评估在这些任务上的进展,程序员需要频繁的反馈,以及判断所观察到的反馈是否将他们引向正确方向的方法。收集、比较和分享反馈通常通过临时手段完成:依赖记忆来比较输出、代码注释或手动截图。为了解决这个问题,我们设计了 Exploriants:一个基于示例的实时编程的扩展。Exploriants 允许程序员放置变点。它收集在探针中捕获的输出,并将其呈现在一个比较视图中,程序员可以自定义该视图以适应其程序领域。我们发现,变点的添加和比较视图鼓励了一种结构化的方法来探索程序的变体。我们通过三个案例研究(图像处理、数据处理和游戏开发)展示了 Exploriants 的能力和适用性。借助 Exploriants,探索式程序员获得了一种直接的方法来评估他们的进展,而不必依赖可能引入错误的临时方法。
英文摘要:
Exploratory programming involves open-ended tasks. To evaluate their progress on these, programmers require frequent feedback and means to tell if the feedback they observe is bringing them in the right direction. Collecting, comparing, and sharing feedback is typically done through ad-hoc means: relying on memory to compare outputs, code comments, or manual screenshots. To approach this issue, we designed Exploriants: an extension to example-based live programming. Exploriants allows programmers to place variation points. It collects outputs captured in probes and presents them in a comparison view that programmers can customize to suit their program domain. We find that the addition of variation points and the comparisons view encourages a structured approach to exploring variations of a program. We demonstrate Exploriants' capabilities and applicability in three case studies on image processing, data processing, and game development. Given Exploriants, exploratory programmers are given a straightforward means to evaluate their progress and do not have to rely on ad-hoc methods that may introduce errors.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。