Toward Rapid Transformation of Ideas into Software
This addresses the problem of slow and difficult software development for a broad range of users as ideas proliferate, but it is incremental as it builds on existing work without introducing new methods.
The paper advocates for making the transformation of ideas into software faster and more accessible as a key long-term goal for software engineering research, and outlines directions to modernize programming technologies to address this challenge.
A key mission of computer science is to enable people realize their creative ideas as naturally and painlessly as possible. Software engineering is at the center of this mission -- software technologies enable reification of ideas into working systems. As computers become ubiquitous, both in availability and the aspects of human lives they touch, the quantity and diversity of ideas also rapidly grow. Our programming systems and technologies need to evolve to make this reification process -- transforming ideas to software -- as quick and accessible as possible. The goal of this paper is twofold. First, it advocates and highlights the "transforming ideas to software" mission as a moonshot for software engineering research. This is a long-term direction for the community, and there is no silver bullet that can get us there. To make this mission a reality, as a community, we need to improve the status quo across many dimensions. Thus, the second goal is to outline a number of directions to modernize our contemporary programming technologies for decades to come, describe work that has been undertaken along those vectors, and pinpoint critical challenges.