── ── Mental model
Deep Work
Cal Newport's framework (2016): deep work = undistracted, cognitively demanding activity that creates hard-to-replicate value; shallow work = logistical, responsive, easy-to-replicate work that fills modern calendars. The most valuable knowledge work is disproportionately the product of deep work; shallow work is increasingly automatable. Three foundations: attention research shows 15-25 min recovery cost per interruption; expert performance requires deliberate practice (structurally…
How it works
Step 1 — Instrument: Track one week. Most workers estimate 4-5 hrs/day deep; measurement shows 0.5-2 hrs in fragments. - Deep work hours/day (tracked): · Shallow hours/day: · Ratio: · Surprise factor: Step 2 — Identify: What specific deep work, if done well, produces the most valuable output? Why isn't it happening? What barriers?
Step 3 — Choose a philosophy: Monastic (eliminate shallow entirely); Bimodal (alternating deep/shallow periods); Rhythmic (fixed daily block — best default for most); Journalistic (switch in whenever a gap appears — requires strong attention discipline, worst for beginners).
Step 4 — Build the practice: Same time + location + setup (ritual). Calendar the block, mark "Do not interrupt." End with a shutdown phrase to prevent cognitive leakage. Boundary is binary: Slack/email closed or you are not in deep work.
When to use it
- user says 'I can never find time for focused work', 'I feel busy but produce nothing', 'I keep getting interrupted', 'I want to protect my focus time', 'flow state', 'deliberate practice', 'Cal Newport', or asks how to make progress on hard cognitive work (research, writing, engineering, strategy)
When not to use it
the role is genuinely high-coordination with no discrete focus windows (ER medicine, real-time trading desk), or user is in a short-term emergency where availability must dominate.
Worked example
Donald Knuth and the No-Email Policy, 1990
The most extreme and well-documented case of deep-work practice in modern science is Donald Knuth, Professor Emeritus of Computer Science at Stanford and author of The Art of Computer Programming (TAOCP), the multi-volume work he has been writing since 1962 and which Bill Gates once called "the best book on programming ever written."
Install this skill (free, MIT)
npx skills add deciqAI/knowledge-skills