Research has argued that “ordinary conversation” and its organization are foundational to “institutional” talk (Drew & Heritage, 1992), and that institutional forms can be characterized as constraints on such a sequential organization. Such an argument can be extended to technology-mediated interaction, in which participants may orient jointly to “standard” conversational sequences and the technology’s interactional “affordances” to achieve interactional ends. We discuss here how such an interactional goal (closing a technology-mediated encounter) may be achieved in live video streaming. We argue that while participants can be seen to do that in a way that orients to the organization of closings in ordinary conversation, they do so in a way that is sensitive to the affordances of live video streams. We find that: a) Most of the streams involve closing sequences, thus framing the live stream as a social encounter for which closing is relevant; b) A partial and relaxed orientation to talking heads configuration in live stream closings; c) A four-part closing sequence; d) Two different topic development in closing the live streams—a topic message comes from the audience and some “mentionable noticeables” initiated by the streamer.

研究认为，“普通谈话”及其组织是“制度性”谈话的基础（Drew & Heritage，1992), 并且制度形式可以被描述为对这种顺序组织的约束。这样的论点可以扩展到以技术为媒介的交互，其中参与者可以共同定位到“标准”对话序列和技术的交互“功能可供性”以实现交互目的。我们在这里讨论如何在实时视频流中实现这样的交互目标（结束以技术为媒介的相遇）。我们认为，虽然可以看到参与者以一种面向普通对话中结束语组织的方式来做这件事，但他们这样做的方式对实时视频流的可供性很敏感。我们发现：a）大多数流都涉及关闭序列，因此将直播流构建为与关闭相关的社交相遇；b) 直播结束时对头像配置的部分和放松的定位；c) 一个由四部分组成的结束序列；d) 关闭直播的两种不同的话题发展——话题信息来自观众和主播发起的一些“mentionable noticeables”。