| name | cognitive-rhythm-writing |
|---|---|
| description | 説明的な文章に緩急を設計するための規範。緩急を装飾ではなく認知モードの切替（観察→逡巡→断定→再観察）と未回収の緊張の管理として扱い、文の拍、段落の密度波形、節の入り方、緩みと駄文の判別、執筆後の機械的な点検手順を定める。読み物として読ませたい章・記事・解説文を生成するとき、または「密度はあるが平坦でおもしろくない」文章を診断・修正するときに使用する。 |
密度の高い文章が退屈になるのは、情報が多いからではなく、全文が同じ認知モードで書かれているからである。 この規範は、読者の認知モード（観察する、迷う、確信する、確かめ直す）を意図的に切り替え、常に「続きを読む理由」を維持することで、読み進める推進力を作る。
A pattern for building personal knowledge bases using LLMs.
This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.
Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.
Every element in current Baseline
browsers
that would need a margin or padding removing in a reset stylesheet:
| Element | Chrome | Chrome for Android | Edge | Firefox | Firefox for Android | Safari | Safari for iOS |
|---|---|---|---|---|---|---|---|
body |
✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ |
h1 |
✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ |
| // Types for the result object with discriminated union | |
| type Success<T> = { | |
| data: T; | |
| error: null; | |
| }; | |
| type Failure<E> = { | |
| data: null; | |
| error: E; | |
| }; |
jsconf2024 で発表したスライドのソースコードだが、テキスト版のが読みやすい気がしたので公開する。 画像のデッドリンクは面倒なので、気が向いたら…
Someone asked me about the spacing scale in Braid and I thought I'd share my thoughts here. For reference this is what they were referring to: https://github.com/seek-oss/braid-design-system/blob/ecdd7f3060cab2a6675cde7dd45bcd510caa268b/packages/braid-design-system/src/lib/themes/baseTokens/apac.ts#L166-L175
I could maybe have turned this into a blog post, but I'm just dropping my rough thoughts here instead.
In practice, the T-shirt size scale felt like indirection because our designers didn't think in those terms so it actually got in the way of developers and designers communicating.
The names also weren't really semantic in the sense that we couldn't confidently change them later without breaking a lot of layouts.
It also made it difficult to name new values to the scale, e.g. if we needed to add a value between "small" and "medium", what would we call it? I've seen things like "xmedium" in other systems before which shows how awkward this can get.
| import gzip | |
| def gzip_search(query: str, candidate_chunks: list[str], top_k: int=1): | |
| """ | |
| 文字列ベースで類似したテキストチャンクを推定するアルゴリズム. | |
| `query`, `chunk`, および`query + " " + chunk`をそれぞれgzipで圧縮し、編集距離のようなものをベースに評価する. | |
| Parameters: | |
| query (str): 検索クエリとして使用する文字列. | |
| top_k (int, optional): 返される類似チャンクの上位k個を指定する (default: 1). |
| // Turn all HTML <a> elements into client side router links, no special framework-specific <Link> component necessary! | |
| // Example using the Next.js App Router. | |
| import { useRouter } from 'next/navigation'; | |
| import { useEffect } from 'react'; | |
| function useLinkHandler() { | |
| let router = useRouter(); | |
| useEffect(() => { | |
| let onClick = e => { |
| /** | |
| * @type {import('prettier').Options} | |
| */ | |
| module.exports = { | |
| plugins: [require.resolve('prettier-plugin-astro')], | |
| overrides: [ | |
| { | |
| files: '**/*.astro', | |
| options: { parser: 'astro' } |