Notes on 乘物游心录 · Technical Notes

Understanding is not storing information

Mon, 20 Jul 2026 09:00:00 +0800

After reading a passage, it is easy to mistake recall for understanding. Repeating the author’s conclusion is useful, but it mainly demonstrates successful storage and retrieval. The words may be available while the argument remains foreign.

Return from the conclusion to the problem
#

One test of understanding is to set the conclusion aside and ask what problem the explanation was designed to solve. Its value lies not only in what it asserts, but also in the tempting alternatives it rules out.

Why efficiency can create more waste

Sun, 12 Jul 2026 09:00:00 +0800

Efficiency is usually treated as an unquestioned good: machines should not sit idle, people should not wait, and every minute should be used. In an interdependent system, however, keeping every local resource busy often moves waste instead of removing it.

Activity is not throughput
#

When one step finishes work that the next step cannot accept, the apparent output becomes inventory, queues, and coordination overhead. The utilization report improves while the result customers need arrives no sooner.

The history of technology is first a history of organization

Sun, 28 Jun 2026 09:00:00 +0800

Technology history is often told as a sequence of inventions: a laboratory makes a breakthrough, a company launches a product, and a performance curve rises. Technologies that reshape society rarely travel on an invention alone. They must pass through manufacturing, standards, capital, skills, and supply chains before they become dependable infrastructure.

Leave a place for uncertainty

Tue, 19 May 2026 09:00:00 +0800

Project plans often express commitment through a single date, budget, and outcome. The danger is not commitment itself. It is the implication that sufficient detail can make uncertainty disappear. Deviations then look like execution failures, bad news gets hidden, and the plan loses its value as a decision tool.

Good design works before the error

Fri, 03 Apr 2026 09:00:00 +0800

Many systems treat error handling as a clearer message shown after failure. That is better than silence, but it still arrives after the mistake. Better design moves forward in time and helps people understand their options and consequences before acting.

Make possible actions visible
#

When an interface shows outcomes but hides state, users must guess at the next step. Visibility means showing where they are, what can be done, which actions are unavailable, and what will happen after completion.

Break large problems into repeatable small ones

Mon, 16 Mar 2026 09:00:00 +0800

“Break the big problem into smaller ones” is common advice. But a longer task list does not automatically reduce complexity. Dependencies can remain tangled, information can disappear at hand-offs, and every small task may still be a one-off that must be reinvented next time.