Orders of Magnitude
summary
Human number sense ends at small counts so number sense ends at four and naming numbers doesn’t mean understanding them; concrete scales are given (hair ~1e-4 m, pillar of creation ~1e16 m, electron ~1e-15 electron width, CPU cycles ~1e-13 s). Practical takeaways: accept long Unicode names and weird emails (don’t rely on brittle regex, send a confirmation email), harden code for concurrency (ticket races, duplicate charges) with use mutexes and DB transactions or background jobs, and design for the billionth request.
AI-written summary · about
chapters (25)
- 0:00Intro
- 1:26Humans have number sense...
- 3:13Many different names for number words
- 3:43Lie: Counting things is easy
- 3:53Counting takes mental energy
- 4:09Roman
- 6:51Lie: Naming numbers allows us to understand them
- 9:04test 'a good test' do setup test assert test.is_good end
- 10:27Human Hair E-4
- 10:43Mountains E4
- 11:14Blink E-4
- 11:45Thousandths to Thousands
- 12:14Nimrud Lens, 750 BC
- 13:25Expanding World
- 13:53Microprocessor memory cell
- 14:24Electron E-15
- 14:58Pillars of Creation E 16
- 15:48Processor Cycle E-13
- 16:01Bristlecone Pine
- 16:21Lie: We've been able to explore the world in great detail for a long time
- 16:38Estimating the Odds
- 17:41Lie: Our brains are good at calculating odds
- 20:28Non-standard characters
- 21:56Database Transactions
- 23:34The edge case is a certainty
related talks
23mDesigning APIs for humans
41mWhat's in a name?
29mNaming Made Easier
01h 17mNaming is Hard: Let's Do Better
01h 24mUnicode: From First Principles to Practical Implementation
10mLightning Talks - What's in a name?
23mDeciphering Alerts – Why is it so hard to make sense
52mThere's No Such Thing As Plain Text