图形学与人机交互2022dougengelbart.org
Doug Engelbart 面向知识型组织的设计（1992）
Doug Engelbart’s design for knowledge-based organizations (1992)
入门 · 约 45 分钟读完前置知识：知识管理基础、超文本/Memex 概念、组织设计基础
导读
这篇 1992 年的文章/演讲材料讨论如何为知识型组织设计支持结构：Engelbart 把组织看作多个相互作用的知识域，关注知识从采集、加工到保存和应用的过程，并提出 CODIAK 与按时间相关性分层的方法。评论将其视为今日 Obsidian、Foam、PARA 等知识管理工具的先声。
HN 关注点：为何像对 Obsidian/Foam/PARA 以及 HN 自身知识库的预言，并关注单一树状视图的局限。
HN 评论精选
这几条评论里：质疑 1 补充 1
补充HN 本身是单一树状、特权视图的知识库；Memex/Engelbart 指向更强的多视图超文本，需要更好机制。
Bush 的 Memex 和 Engelbart 的“所有演示之母”暗示的东西，比我们最终得到的 HTML 和过早优化强大得多。读 Jtsummers 链接的 PDF 时，我回想以前支持过的销售，有人收件箱里有 20+ GB 邮件，把我逼疯（当时 Exchange Server 处理得很糟）；现在我知道原因了：他们是在搭一个可搜索的超链接/线索化数据库。HN 也在随时间做同样的事，构建保留上下文、可搜索的树状数据库。主要限制是只有一种可能的树，而且是特权视图。这里至少沉淀了十亿美元级的知识，却要靠 dang 和强大社区才能让这个视图保持理性，因为人性中的腐化力量和垃圾/诈骗的盈利动机。一定会有更好的办法。
mikewarot查看英文原文
Bush's Memex and Engelbart's mother of all demos hint at something much more powerful than HTML and the premature optimization that we got as a result. In reading the pdf linked by Jtsummers, I recall how some of the sales people I supported had 20+ gigabytes of email in their inboxes... which drove me nuts (Exchange Server really didn't cope well with it back then), and now I know why, they were building a hyperlinked database (well, threaded) that they could search through. Here in HN, we do the same thing over time, building a tree structured database that preserves context and is searchable. The main limitation here is there's only one possible tree, and it's the privileged view. There's at least a billion dollars of knowledge captured here in HN. Yet, it takes Dang and a strong community to keep that one view rational, due to the corrupting forces of human nature and profit motives from spam and scams. There's got to be a better way.
质疑该文更像知识管理工作流（CODIAK、时间分层、近 PARA）的预言，而非组织理论；与 Obsidian/Foam 一脉相承。
我把它读作对 Foam、Obsidian.md 之类工具的预言。参见 2020 年关于 Foam 的讨论。尽管 PDF 标题是“为知识型组织设计”，它更多讲的是从知识捕获到使用再到修剪的组织流程与方法能力，而不是组织本身在知识上的工作。文中只有一段旁白谈组织：Engelbart 把每个组织视为多个相互作用的知识域，研究重点是为这些知识域内部和跨域的知识采集与提炼设计支持结构。随后它跳进某个知识域，提出该域内的知识管理方法。有趣的是，这预言了在“风格压过语义”的两个黑暗十年后，知识管理工具近来的合流。第一个想法是知识引入、加工和保存，他称为 CODIAK。为使其可行，他提出按时间相关性分层的方法，很像今天 KM 工具用户采用的 PARA。之后讲功能影响，并描述了几乎……（评论原文在此截断）
Terretta查看英文原文
I read this as a prophecy of Foam (https://github.com/foambubble/foam), Obsidian.md (https://obsidian.md/), and the like. See this thread on Foam from 2020: https://news.ycombinator.com/item?id=23666950 Despite ”design for knowledge based organizations” being the title of the article in the PDF, this is perhaps more about approach and capabilities for organizing flow from capture to use to pruning of knowledge, more than the organizations working on knowledge. There is one aside about the organization itself: > Engelbart sees every organization as a collection of interacting knowledge domains. He has focused his research on designing support structures for knowledge collection and refinementwithin and across these knowledge domains. But then that jumps down into a given domain, and goes on to propose a knowledge management approach within that (and each) domain. This is where it gets interestingly predictive of the recent coalescence in knowledge management tools emerging after the two dark age decades of style over semantics. The first idea is a concept of bringing knowledge in, working it, and keeping it. He calls this Concurrent Development, Integration, and Application of Knowledge (CODIAK) process. To make this workable he proposes a time-relevance layered approach very similar to the (perhaps easier to action) “PARA” method adopted by many KM tool users today: https://fortelabs.com/para After this, he goes into functional implications, and describes almost to a T capabilities in the latest round of knowledge management tooling “systems” such as, say, Obsidian.md, built on standards (markdown, wikilinks, front matter) and extensibility enabling journaling, querying, views, as first class citizens. When hypertext markup iterated into a page style description advertorial tool rather than linked semantic knowledge structure, that left an opening for the tools we’re seeing now. Today, Obsidian.md is close to the mark, except, of course, for approachability by casuals. For casuals, consider e.g. Craft.do (https://www.craft.do/). If you’re doing a comparison before jumping in, it’s worth diffing the tools against Englebart’s take. I’d argue he’s dead on.
相关资源
从这篇论文所有 HN 讨论帖的全部评论里整理出来的链接。
- 代码
- github.com/foambubble/foam被提到 2 次
历次讨论
- 2022Hacker News 讨论帖147 分 · 30 评
同领域还值得读
像素不是小方块 (1995)
A Pixel Is Not A Little Square (1995)
这篇 1995 年的经典备忘录讨论数字图像中像素应如何理解：像素不是有面积的小方块，而是对连续图像的采样点，重建连续图像需要合适的重建滤波器。
图形学与人机交互2014 · 372 分 · 241 评进阶 · 约 20 分钟编辑精选自动化的讽刺（Ironies of Automation，1983）
Ironies of Automation (1983)
这篇 1983 年的经典论文讨论自动化（尤其过程控制）中的悖论：自动化越先进，人类操作员在异常处理和人工接管中的作用可能越关键；自动化常把常规任务拿走，留下监控、诊断和罕见故障处…
图形学与人机交互2019 · 147 分 · 38 评入门 · 约 30 分钟编辑精选GDPR 同意框中的暗黑模式
Dark patterns in GDPR consent boxes
研究 GDPR 同意框中的暗黑模式，即网站如何通过界面设计引导用户同意 Cookie 与追踪。
图形学与人机交互2020 · 328 分 · 122 评 · 引用 6进阶 · 约 30 分钟用稀疏条带在 CPU 上实现高性能 2D 图形渲染（硕士论文）
High-performance 2D graphics rendering on the CPU using sparse strips
这篇硕士论文研究如何在 CPU 上实现高性能的 2D 图形渲染，核心思路是用“稀疏条带（sparse strips）”结构来表示和光栅化路径，以降低内存占用并提升渲染速度。
图形学与人机交互2025 · 281 分 · 35 评专业 · 约 90 分钟GDPR 后的暗模式：同意弹窗及其影响
Dark patterns after the GDPR: consent pop-ups and their influence
研究 GDPR 生效后网站 consent pop-ups 中的 dark patterns，以及这些弹窗如何影响用户的同意选择。
图形学与人机交互2021 · 251 分 · 228 评进阶 · 约 60 分钟