Productivity & Learning — devopsmantra wiki

577 saved links, 2013–2026 (peak: 2023). This is your longest-running tag and the most personal one: a decade-plus subscription habit to Ness Labs and Farnam Street, a note-taking thread that runs from Roam through Obsidian to LLM wikis — the vault you’re reading right now is that thread’s current chapter — and a recent pivot to the question every save from 2025 onward circles: what does productivity even mean when the agent does the typing? Less about hacks than about learning how to learn, which turned out to be the right bet for a career that reinvents itself every few years.

Learning how to learn — the meta-skill#

The deepest vein in the tag, and the one that made your DevOps-to-AI transition feel routine rather than terrifying.

Barbara Oakley: Learning How to Learn (Talks at Google) — “I’ve had this queued up for weeks… It is AWESOME!” Your own verdict; start here.
The Feynman Technique — the fs.blog classic — and the podcast version you flagged #mustKnow.
Ness Labs: learning how to learn · why “practice makes perfect” is only half true · self-education as superpower · unbounded learning.
Retention, the leaky bucket problem: the Ebbinghaus forgetting curve · how to remember what you read · the Ness Labs take on the same.
Spaced repetition, then and now: Effective Spaced Repetition · build your own Anki decks · FSRS: spaced repetition got way better thanks to machine learning — the old thread meeting your new field.
How to remember everything you learn (video) and the RAIL framework for learning complex skills quickly.
Reading as a skill: Mortimer Adler’s How to Read a Book · the Ness Labs distillation · writing to think — “a space to practice reasoning.”
How to learn things at 1000x the speed · learn almost anything in 5 minutes and remember it forever · effective studying: explain aloud, weed out distractions.

Mental models — the Farnam Street shelf#

You’ve been collecting Shane Parrish and Charlie Munger for nearly a decade. These are the keepers.

The Great Mental Models — the book series; first principles, second-order thinking, probabilistic thinking.
First-principles thinking — “good @farnamstreet blog post,” per your note — and Musk’s semantic tree: trunk before leaves.
Munger: adding mental models to your toolbox and Munger’s psychology of misjudgment speech — “probably the best thing to read when you’re starting to invest.”
The work required to have an opinion · Bertrand Russell on avoiding foolish opinions · open vs. closed-minded people.
Good thinking requires time · explore or exploit? · the difference between amateurs and professionals.
Hamming’s “You and Your Research” — “knowledge and productivity are like compound interest.”
Howard Marks: how to think about risk (video) and Carol Dweck’s growth mindset.
10 mental models for learning anything — where the two threads of this page meet.

Note-taking & the second brain — the thread that became this vault#

Roam (2020) → Obsidian (2023) → LLM wikis (2026). This page is the practice, not just the reading about it.

The Zettelkasten roots: Luhmann’s note-taking habit · no, Luhmann did not invent the Zettelkasten · from note-taking to note-making.
Commoncog: how note-taking helps you rapidly develop expertise — “best article I’ve read on this”; the conceptual case for backlinks and blocks.
The Ness Labs playbook: a taxonomy of notes · taking notes while reading · interstitial journaling.
The Roam era: Roam as a mind-mapping tool and Magical Academic Note-Taking, now free.
The Obsidian era: Obsidian vs. Heptabase, compared · a full math textbook built in Obsidian · Obsidian Bases: notes become databases · connecting Zotero and Obsidian.
The LLM-wiki era — the 2026 cluster that describes what you’re doing here: an example LLM wiki built on Obsidian · why LLM wikis need a visual layer · a second brain that works like an LLM wiki for agents · Claude Code scaffolding with Obsidian integration · Claude Code second-brain middleware · a PKM system with AI agents and hooks.
10 NotebookLM use cases as a second brain and Rocketnotes: markdown notes with a serverless RAG pipeline.

Focus, procrastination & the anti-hustle turn#

The early saves chase morning routines; the later ones question the whole premise. The arc is visible.

A Smart Bear on focus — stop doing so many things — and “deep work is statistically near-impossible” — focus has been engineered out of the workday.
you wildly overestimate your focused hours · recognize the value of your attention and allocate it consciously · it’s time to regain your ability to focus.
Procrastination, honestly: 4 research-backed secrets to stop procrastinating — your note: “I put off reading this for a week.” · pinpoint the exact cause of your procrastination · 5 lessons from Tim Pychyl’s book (video) · Neil Gaiman’s rule: be bored or write.
Habits and systems over goals: habits vs. goals · forget goals, focus on systems · implementation intentions — the trick that doubles your odds · making “the right thing” feel effortless.
The frameworks, when you want one: Ness Labs’ primer on GTD, Pomodoro, time blocking, Kanban and friends · the Eisenhower matrix · flow state.
The counter-current: “toxic productivity” and how to end it · from FOMO to JOMO — “lovely read ☺️” · Ferriss on overwhelm: “if this were the only thing I accomplished today, would I be satisfied?” · Internal Family Systems for doing better work while less overwhelmed.
Inside Linear: taste, craft, and focus — what focus looks like at company scale.

Productivity in the AI era#

The 2024–2026 saves hold two ideas in tension: AI is the biggest productivity unlock of your career, and the measured gains keep coming back complicated.

Developer productivity in the age of generative AI: a psychological perspective — “a colleague of mine wrote his master’s thesis on this.”
The skeptical evidence: DORA: AI improves flow and satisfaction but paradoxically cuts time on what developers consider productive · Upwork study: gen AI increases workloads · vendor productivity numbers have a dozen asterisks — test in your own environment · the Great AI Deskilling has begun.
“Don’t confuse code generation with productivity” — productivity is what happens after generation: validation, deployment, rollback.
The playbooks: Jeremy Utley’s 5-step playbook — role assignment, context engineering, emphasis · four AI-native mental models: producer → manager, implementation → intent · the agentic coding loop that matters.
The toolbelt: a marketplace of Claude subagents, skills, and plugins · thousands of MCP servers and agent skills, categorized · Matt Pocock’s /teach skill — learn anything via an agent.
The mindset: how to not feel constantly behind — compete with AI and you lose; use it for others and you find space · human curiosity in the age of AI · how learning to code has forever changed: build things, ask the agent.
Measuring it properly, pre- and post-AI: Fowler: qualitative developer-productivity metrics, with a sample survey · how Google measures developer productivity internally.

Reading lists & curricula#

You collect other people’s syllabi the way some people collect stamps.