"The average user doesn't give a damn what happens, as long as (1) it works and (2) it's fast"
About this Quote
A mathematician famous for security-minded engineering doesn’t offer this as a shrug; it’s a warning shot. Bernstein’s line strips away the romance technologists project onto “users” and replaces it with a blunt market reality: most people don’t want to understand systems, they want outcomes. “Doesn’t give a damn” is doing deliberate work here. It’s not contempt for users so much as contempt for the fantasy that users will rescue bad design with patience, curiosity, or loyalty.
The parenthetical numbering reads like an engineer’s spec, but it’s also satire: after all the moralizing about transparency, elegance, and “best practices,” the acceptance test is brutally small. Works. Fast. Everything else - purity, ideology, even many security properties if they’re invisible - becomes negotiable unless it affects those two metrics.
In context, this lands inside Bernstein’s long-running argument with the software world about incentives. Security and correctness are hard to sell because their success is the absence of catastrophe; speed and functionality are felt immediately. That mismatch explains why “good enough” systems win, why UX polish beats formal guarantees, and why engineers repeatedly relearn the same lesson after each breach: users weren’t voting for architecture, they were voting for frictionlessness.
The subtext for builders is uncomfortable and practical: if you want people to care about safety, privacy, or reliability, you can’t ask them to. You have to bake it in without slowing things down or breaking what already “works.”
The parenthetical numbering reads like an engineer’s spec, but it’s also satire: after all the moralizing about transparency, elegance, and “best practices,” the acceptance test is brutally small. Works. Fast. Everything else - purity, ideology, even many security properties if they’re invisible - becomes negotiable unless it affects those two metrics.
In context, this lands inside Bernstein’s long-running argument with the software world about incentives. Security and correctness are hard to sell because their success is the absence of catastrophe; speed and functionality are felt immediately. That mismatch explains why “good enough” systems win, why UX polish beats formal guarantees, and why engineers repeatedly relearn the same lesson after each breach: users weren’t voting for architecture, they were voting for frictionlessness.
The subtext for builders is uncomfortable and practical: if you want people to care about safety, privacy, or reliability, you can’t ask them to. You have to bake it in without slowing things down or breaking what already “works.”
Quote Details
| Topic | Technology |
|---|
More Quotes by Daniel
Add to List
