Soylent is a word processing interface that integrates crowdsourced human contributions to aid in complex writing tasks such as error prevention, paragraph shortening, and automation of tasks like citation searches and tense changes. The interface consists of three main components: Shortn, a text shortening service; Crowdproof, a human-powered spelling and grammar checker; and The Human Macro, an interface for offloading arbitrary word processing tasks. The paper introduces the Find-Fix-Verify crowd programming pattern, which splits tasks into generation and review stages to improve worker quality. Evaluation studies demonstrate the feasibility of crowdsourced editing and investigate reliability, cost, wait time, and work time for edits. The authors discuss the challenges and limitations of using Mechanical Turk workers, including high variance in effort, the introduction of errors, and the need for clear guidelines to prevent buggy commands. The paper concludes by exploring issues such as wait time, cost, legal ownership, privacy, and domain knowledge in the context of interface outsourcing.Soylent is a word processing interface that integrates crowdsourced human contributions to aid in complex writing tasks such as error prevention, paragraph shortening, and automation of tasks like citation searches and tense changes. The interface consists of three main components: Shortn, a text shortening service; Crowdproof, a human-powered spelling and grammar checker; and The Human Macro, an interface for offloading arbitrary word processing tasks. The paper introduces the Find-Fix-Verify crowd programming pattern, which splits tasks into generation and review stages to improve worker quality. Evaluation studies demonstrate the feasibility of crowdsourced editing and investigate reliability, cost, wait time, and work time for edits. The authors discuss the challenges and limitations of using Mechanical Turk workers, including high variance in effort, the introduction of errors, and the need for clear guidelines to prevent buggy commands. The paper concludes by exploring issues such as wait time, cost, legal ownership, privacy, and domain knowledge in the context of interface outsourcing.