Maggie Appleton 写设计、人类学和智能体界面时,习惯先画清楚隐喻,再谈工具。把「Ambient Co-presence」整理成可阅读的中文笔记:问题在哪、界面默认会把人带去哪、落地时该改什么。原站导航、评论回响和广告已去掉。

The Context

It’s hard to create a subtle sense of shared, synchronous space among multiple people on the web.

You and I are currently in an a synchronous exchange. I’ve written these words in the past. You’re reading them at a point in the future. We are not in sync.

The Pattern

Despite being on the same web site – which suggests a shared digital space – it bears very little resemblance to how we might interact in a shared physical space in real-time.

We currently have no visual, audible, tactile, spatial, or embodied awareness of one another. We also have no awareness of the other people reading this post, even if they’re doing it at the exact same moment.

Multiplayer Cursors

For a long time, the asynchronous nature of the web was a technical limitation. Internet speeds and bandwidth simply couldn’t accommodate anything beyond sending reams of static text to one another with long, slow lags in between.

But things have progressed. We now take for granted that our web connections refresh every few milliseconds. We experience most of it in real-time; spicy Twitter drama, collaborative editing in Google docs, overcrowded Zoom meetings, sensationalist BBC breaking news coverage, World of Warcraft tournaments, and the flood of emoji reactions on Twitch livestreams load in the blink of an eye.

Spatial Audio

But it feels as though we haven’t quite clocked what’s possible here. The synchronous, multiplayer web is brand new, and the design space around it is still relatively unexplored.

Most of the web is still designed to accomodate slow-loading, single-player experiences. We all stare into linear, individualised feeds that lack the qualities of shared physical space; directionality, subtle awareness of those around us, opportunities for ad-hoc interactions, freedom to explore, and most critically, a shared understanding we are seeing the same reality as people in our immediate vicinity. There is no ‘ vicinity ’ when you are alone starin

Annotations and Trails

We don’t necessarily need to constantly interact with people “around” us on the web. The sensation of being in the quiet companionship of someone else, like reading next to them in a cafe, is what we’re missing. The sense of ambiently sharing space – of being co-present – while engaged in other activities is a staple of shared public spaces that we’re still figuring out how to design in the digital realm.

Our current “multiplayer” experiences draw too much attention to the multiplayer-ness. The other people around you demand attention. They move. They flash. They point to exactly what they’re focused on, drawing you away from your own focal point. We are missing out on a fuzzier, softer sense of the shared web.

References and Further Reading

Ambient co-presence is the sensation of sharing a space or context with other people, without directly interacting or continuously communicating with them. You’re all aware of each other through peripheral vision, ambient noise, and embodied sensations but aren’t necessarily doing the same tasks or directing your attention to the same things.

This looks like quietly reading on the back porch with family, sitting in a cafe full of people softly chatting or tapping away at laptops, or people watching from a shabby seat on the London underground, marvelling at the vast array of people that can exist in a single space.

落地时建议先做的 5 件事

  1. 先写清这个工具在强化思考还是在替人思考。
  2. 默认交互不要只会谄媚:该追问、该给反例、该要求证据。
  3. 智能体界面要露出推理步骤,而不是只给一个光滑答案。
  4. 知识发布用可生长的笔记,而不是一次性营销长文。
  5. 改产品前先画隐喻:用户以为自己在做什么,系统实际在做什么。

和智能体产品怎么接

龙虾PRO做 OpenClaw 落地时,最该从这类笔记里拿走的是「别把聊天框当唯一界面」。数字员工要能追问、能验证、能把过程摊开,而不是只负责说好话。

本文侧重全链路风控方法论。落地时请用自身业务单据做回放验证,不要把示例阈值直接当生产策略。 相关:风控体检 · 方案资源

常见问题 FAQ

什么是AI智能系统?

「AI智能系统」可概括为:It’s hard to create a subtle sense of shared, synchronous space among multiple people on the web. 本文从定义、方法与实践要点展开说明。

为什么要关注AI智能系统?

关注AI智能系统,是因为它直接影响效率、风险与可复制性。文中指出:It’s hard to create a subtle sense of shared, synchronous space among multiple people on the web.

如何落地AI智能系统?有哪些关键步骤?

建议按以下路径推进AI智能系统:1) 先写清这个工具在强化思考还是在替人思考。;2) 默认交互不要只会谄媚:该追问、该给反例、该要求证据。;3) 智能体界面要露出推理步骤,而不是只给一个光滑答案。;4) 知识发布用可生长的笔记,而不是一次性营销长文。;5) 改产品前先画隐喻:用户以为自己在做什么,系统实际在做什么。。细节见正文对应章节。

AI智能系统适合哪些人或团队?

AI智能系统更适合:产品/技术负责人、运营与增长团队、需要落地智能体或自动化的中小团队、关注「AI智能系统」方向的读者。若你只需要单次聊天式问答,可先读概念;若要上生产,请重点看步骤、权限与风控相关段落。

关于「The Context」,本文给出了什么结论?

在「The Context」部分,要点是:ese words in the past. You’re reading them at a point in the future. We are not in sync. The Pattern Despite being on the same web site – which suggests a shared digital space – it bears very little resemblance to how we

关于「The Pattern」,本文给出了什么结论?

在「The Pattern」部分,要点是:e. We currently have no visual, audible, tactile, spatial, or embodied awareness of one another. We also have no awareness of the other people reading this post, even if they’re doing it at the exact same moment. Multipl