SHF: Small: Human-Centered Software Synthesis

SHF：小型：以人为本的软件综合

• 批准号: 1116362
• 负责人: Armando Solar-Lezama
• 金额: $ 40.48万
• 依托单位: Massachusetts Institute of Technology
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2011
• 资助国家: 美国
• 起止时间: 2011-09-01 至 2015-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1116362&HistoricalAwards=false
• 关键词: SHF; Small; Human; Centered; Software

The goal of this project is to develop new forms of programming that are more intuitive and more flexible than traditional programming models. Our new programming model is targeted towards low-level system's programming for systems like databases and operating systems where performance and reliability requirements make development particularly challenging. One of the components of our new programming model is new set of graphical notations to allow programmers to pictorially describe the behavior of their programs and automatically get implementations from these high-level drawings. The new programming technologies developed and released as part of this project could have a dramatic impact in programmer productivity, particularly for low-level systems programming. The technology we are developing relies on new algorithms for constraint based program synthesis that allow the system to combine different forms of input ranging from high-level graphical descriptions of program behavior with more formal notations such as invariants and pre/post conditions. The project will also explore new forms of interaction between the software synthesis engine and the programmer, allowing for a more interactive development model for complex algorithms. These new forms of interaction will be analyzed and evaluated in terms of their impact on programmer productivity through user studies and other forms of usability evaluation.

这个项目的目标是开发新的编程形式，比传统的编程模型更直观，更灵活。我们的新编程模型针对的是数据库和操作系统等系统的低级系统编程，这些系统的性能和可靠性要求使开发特别具有挑战性。我们的新编程模型的组件之一是一组新的图形符号，允许程序员以图形方式描述其程序的行为，并自动从这些高级绘图中获得实现。作为该项目的一部分开发和发布的新编程技术可能会对程序员的生产力产生巨大影响，特别是对于低级系统编程。我们正在开发的技术依赖于新的算法约束为基础的程序合成，使系统联合收割机不同形式的输入范围从高层次的图形描述的程序行为与更正式的符号，如不变量和前/后条件。该项目还将探索软件合成引擎和程序员之间的新的交互形式，为复杂算法提供更具交互性的开发模型。将通过用户研究和其他形式的可用性评估，分析和评估这些新的交互形式对程序员生产力的影响。