Sébastien Dubois

Sébastien Dubois

@dsebastien.net

Organize your thoughts, life & work⚡ AI, Obsidian & Knowledge Management. Author. Coach. Entrepreneur. Building AI + Obsidian tools 💌 https://dsebastien.net

513 followers221 following8.8K posts

View full thread
View full thread

Too many results is a reason to move to the next step, just like zero results. Name, then phrase, then filters. Stop as soon as you find the note.

View full thread

Only a word or a phrase? Open the Search view (Ctrl/Cmd+Shift+F). "exact phrase", -exclude, path:, tag:, [property:value], task-todo:, /regex/.

Six kinds of work, six ways they end: - Idea: promoted or dropped - Goal: achieved or abandoned - Plan: applied, revised or rejected - Project: completed or abandoned - Activity: retired, deliberately - Task: done Mix them in one list and nothing ends cleanly.

Assets are activities. Changes to them are projects. Your product, your website, your team's tooling: activities. You keep them alive. A new version, a migration, a redesign: projects. They end. The project ends; the asset carries on.

The out-of-scope list is the easiest part of a project to skip. It's also the one that protects you when someone asks for "just one more thing". That's how scope creep starts. Write down what you're NOT doing before you start.

GTD sets a very low bar: anything that takes more than one action is a project. That's why GTD project lists get so long. I add two requirements: a finish line and a why. No finish line? It's an activity. No why? Ask whether it deserves your time at all.

Use folders for note types. Leave topics to tags and links. Why? A note lives in ONE folder, but it can have as many tags and links as you want. Part 3 is out: folders, the archive, ISO dates, daily notes as a timeline, and a capture-time checklist 👇

View full thread

Every project needs seven things: a finish line, deliverables, a why, a scope (in AND out), one owner, a status and a next action. Put them in one home per project. When anyone asks where it stands, the answer should be one click away.

A task takes a day or less. An idea isn't committed yet. A goal is an outcome you want to be true. An activity never finishes: it has a cadence and a standard instead of a deadline.

My rule for the confusing cases: assets are activities, changes to them are projects. A product you keep alive is an activity, and each of its releases is a project. The release ends. The product doesn't.

The project one-pager I use: - Objective - Success criteria (checkable) - Deliverables - Why (the goal or activity it serves) - In scope - Out of scope - Owner (one person) - Status - Next action

"The team" is not an owner. Every project needs one person accountable for getting it done. Not a group, not a department, not "we'll figure it out". No owner? Then nobody is going to finish it.

If you work inside someone else's project, demand five answers early: - What's the objective and how will we know it's done? - What's in and out of scope? - Who owns it? - Where does it live? - What's my next action?

Put acronyms in parentheses. Name the note "Artificial Intelligence (AI)". Type "AI" or "Artificial Intelligence" in the quick switcher, and it shows up either way. No alias to maintain. Part 2 is out: names, aliases, note types, and the minimum metadata worth adding 👇

Even when they knew details about what they wanted, people in a 2004 MIT diary study used keyword search in only 39% of their searches. We navigate first. We search when navigation fails. That's why findability is decided when you WRITE the note. Part 1 is out 👇

A "Year in Media" page, generated from my notes: - what I watched, played and finished this year - hours per medium - ratings by director and by decade - what's up next Plain Markdown, published on my site. Built from 10,260 notes.

I put the Obsidian search commands and operators worth remembering on one free cheat sheet: quick switcher, Search, Bases, CLI, ripgrep, qmd. Plus the search escalation ladder: which tool to try first, and what to try next 👇 www.dsebastien.net/free-obsidi...

dsebastien.netFree Obsidian Search Cheat Sheet: operators and the search escalation ladderA free 13-page PDF with the Obsidian search operators worth remembering and the search escalation ladder: which tool to try first, depending on what you remember about the note.

My Obsidian search series is free, and the overview is up today. Then one new part every weekday for 12 weekdays, starting tomorrow. Start with the overview. It has the search escalation ladder, the findability stack, and a table that maps what you remember to the tool to use.

The weakness of Obsidian's native Kanban that bothers me the most: it doesn't know which values are valid. Write "In progres" once, anywhere, and the board quietly adds a new column. Your card disappears from the column where you expected it.

I wrote a book on finding notes in Obsidian 🚀 The Obsidian Search Handbook: 254 pages, PDF + EPUB. The science of re-finding, the tools (quick switcher to AI search), the habits, and "Try this" exercises in every chapter. €19 with code LAUNCH until October 29. €29 after.

Pick any task on today's list and trace it upward. Which project or activity does it serve? Which goal does that serve? If you can't answer, either a link is missing, or the task doesn't deserve your time.

I built the Kanban Action Planner because I hit the limits of every Kanban solution I tried. Goals, projects, tasks, books, videos... each note type has its own lifecycle and its own "done". The native boards in Obsidian 1.14 are great. They don't cover that (yet).

An action system rots if you don't look at it. Each level needs its own rhythm: - Tasks: daily and weekly - Projects and activities: weekly and monthly - Goals and plans: monthly to yearly - Ideas: at least once a quarter, or the funnel becomes a graveyard