Why Abstract Models Are the Key to Lifelong Learning

What Should We Learn?

Abstract Models

Zhuangzi said our life is limited while knowledge is endless, making it foolish to chase infinite knowledge with finite life. Therefore, the ultimate goal of learning is not knowledge itself, which is superficial and fleeting, but rather enduring abstractions such as philosophy, methodology, or abstract models.

Mathematical formulas exemplify perfect abstract models, serving as universal tools to reveal the laws governing the universe and nature.

Every discipline possesses its own abstract models, like stars in the sky—some similar, many distinct. Expanding our cognitive structure means expanding the boundaries of these models; the more models we own, the richer our cognition.

What is the immutable essence for computers?

Computer Models

Physically, transistors have two states: on and off; electrically, voltage has high and low levels, corresponding to binary 0 and 1 (excluding quantum computing). Adding more capacitors or lines yields 2⁴, 2⁸, 2¹⁶, etc., and advances in nanotechnology, multi‑core CPUs, and 5G increase the number of representable states.

Regardless of virtual complexity, tracing back to the physical layer starts with capacitors, to binary in mathematics, and to yin‑yang in philosophy.

The computer’s power‑on process resembles a cosmic big bang, launching a journey of bits that travel from disk → bus → memory → CPU, multiplying along the way.

This world is built on a stable philosophical foundation, a mathematical representation of infinite states, and a wave‑particle‑based ultra‑efficient substrate.

Von Neumann Architecture

The classic computer architecture—CPU, memory, controller, I/O—remains stable across PCs, mobiles, and emerging IoT systems, differing mainly in performance and power consumption.

Compiler Principles

Understanding any programming language’s underlying mechanics requires knowledge of lexical analysis, syntax analysis, semantic analysis, regular expressions, and finite state machines—core topics that have hardly changed.

Whether Go, Rust, Java, C/C++, Python, JavaScript, or C#, the focus should be on the compilation process rather than superficial syntax differences.

All languages, regardless of paradigm, construct syntax trees, perform lexical and semantic analysis, and ultimately translate into binary code.

Distributed Systems Principles

Distributed storage systems share a common data replication method, first documented in Lamport’s 1978 paper “The Implementation of Reliable Distributed Multiprocess Systems.”

Despite rapid technological advances, the underlying theory remains unchanged, underscoring the need to reflect, summarize, and solidify knowledge while learning new tools.

This replication principle appears in relational databases (MySQL, SQL Server) and NoSQL stores (Redis, MongoDB), as well as search engines (Elasticsearch) and message queues (Kafka, RabbitMQ).

Methodology

Epistemology of Ignorance

Einstein likened the universe to a watch whose inner workings are unknown to us. Ignorance here means assuming we know nothing, then re‑examining our thoughts. This aligns with Musk’s “first‑principles” thinking.

When systems become bloated and code decays, only a fundamental architectural rethink can temporarily meet business needs—future evolution remains uncertain.

True ignorance is a stance that seeks knowable truths, fostering a sincere learning attitude.

Universal Skepticism

Universal skepticism is not skepticism for its own sake; the universe is a chaotic whole, and disciplinary names are not eternal. Thus, any discipline is a perspective, never absolute.

It separates the self from thought, questioning inherited beliefs. Descartes’ “I think, therefore I am” illustrates this.

Applying universal skepticism, like Einstein did to Newton’s laws, can break collective beliefs and drive innovation.

Thought ≠ Self

When criticized, we react emotionally because we conflate thought with self. As Li Shanchang said, “Thought does not belong to me; it possesses me.” Our thoughts are shadows of ancestral wisdom that merge with us over time.

Thus, we should defend ideas, not ourselves, and welcome correction rather than emotional backlash.

Independent Thinking

Human evolution favored conformity for safety, but blind conformity hampers rational judgment. Modern organizations still exhibit this tendency, threatening independent thought.

Independent thinking involves systematic, abstract, rational analysis—not mere emotional or visual thinking.

In computing, this translates to mathematical thinking: modeling reality numerically to drive industry‑focused computation.

Conclusion

Therefore, our learning goal should be mastering abstract knowledge models—universal keys that unlock various domains—rather than accumulating transient facts. Methodologically, we should employ universal skepticism, epistemic humility, and detach thought from self to continuously challenge and refine our mental models, echoing Buffett’s advice to “break a belief each year, or the year is wasted.”