Thinking at Small and Large Scales

Source and verification limitations (2026-11-02): The post is tagged Wu Jun and Dedao. Only some concepts were located in third-party notes; the full original course was not obtained. The classroom survey about shoes, the programmer anecdote, the dates associated with webpage counts, and comparisons involving banking and pharmaceuticals have not each been verified. These generalizations cannot be treated as established causal explanations for technology replacing traditional industries. Sources and reference links (see the explanation above for the scope of verification): https://geekdaxue.co/read/woniu666@tools_and_notes/google_method_002 ---- Your way of thinking often determines whether you win or lose. People who live in a world of small numbers naturally lack the ways of thinking needed to tackle problems in a world of large numbers. This invisibly places a ceiling on them. Wu Jun once conducted a classroom survey of his EMBA students, asking those who owned 100 pairs of shoes or more to raise their hands. Usually, half the women raised their hands. When he asked those with fewer than 10 pairs to raise theirs, some men usually did. The two groups managed their shoes differently. Those with fewer than ten pairs simply put them together wherever convenient, without needing a system: before leaving, they glanced over them and chose a pair. Almost without exception, those with more than a hundred pairs sorted them into categories. Otherwise, spending 20 seconds considering each of those hundred-plus pairs before going out would take almost an hour, and they would probably be late for their date. Wu Jun draws two concepts from this example: 1. People living in different worlds develop different habits for getting things done. 2. Methods derived from small quantities cannot be applied to problems at higher orders of magnitude. If you live in a world at a smaller scale, your habitual way of thinking is to start with small examples such as 1, 2, and 3. There is nothing inherently wrong with that. But applying the same thinking to large problems often fails. This is common among IT professionals. Wu Jun once saw a director criticizing a young programmer on his team. The director pointed out that the programmer was using a very inefficient sorting method. The programmer replied that there were very few items to sort—only a single-digit number—so any sorting method would take about the same time. Under a tight deadline, he had chosen the method that was easiest to code. For that programmer, living in a world of small numbers naturally meant retaining familiar habits. But once you work in computing, you must follow the rules of the field: you cannot assume in advance that there will always be few values. If the quantity grows, the inefficient method may multiply the computational workload not just tenfold or a hundredfold, but a thousandfold or even ten thousandfold. Without changing that thinking, you will never become a competent computer engineer. The first reason IT professionals hit a career ceiling is that their way of thinking was wrong from the start. No matter how hard the programmer above works at writing code, he may remain a low-level programmer years later. Unless we overcome the limitations of our innate cognition, we cannot grasp the essence of IT, and our careers will not advance. This is why many large companies are killed off by emerging technology companies without even understanding how they died. Today, both IT and non-IT companies offer many examples of businesses failing because they are insensitive to scale or change. The difficulties facing the once-glorious Yahoo are rooted in this. In the early internet, there was relatively little content, so organizing information into categorized directories was effective—just as when someone owns a hundred pairs of shoes or a few hundred books. This gave rise to portals such as Yahoo, which organized information through directories. But in computing, everything grows explosively. Webpage numbers rose from millions in the early 1990s to tens of billions today. Yahoo and other portals therefore declined rapidly, replaced by search companies such as Google. After the advent of Web 2.0, everyone could generate news, putting us in a world of large numbers. Toutiao, which understood computational thinking, therefore approached news distribution very differently from the portals. There is now so much news that grouping it into sections is no longer enough; Toutiao has to filter it in a personalized way. The decline of news portals and the rise of Toutiao are, in effect, a contest between two ways of thinking. Take finance as another example. Jobs in that industry were once considered a “golden rice bowl,” yet today major banks' branch networks are shrinking and lending has become harder than before. The reason is simple: every loan used to be approved manually, and a loan officer might not issue even 10 loans in a week. Experience mattered, but people accumulated it by working things out for themselves, with little exchange among peers. Once Alibaba's Ant Financial appeared, however, the explosive growth in data made loan officers' experience of little use. Ant Financial succeeded because it faced a vast world from the outset and therefore approached problems differently from traditional banks. Thus, although major banks have their own inherent advantages, they face growing threats from Ant Financial and find it difficult to grow their businesses. In traditional pharmaceuticals, analyzing cases and data from more than a thousand people was considered impressive. Yet drug development remained slow, and many prescription medicines were actually of little use to most people. Unlike traditional pharmaceutical companies, Human Longevity and Google's big-data healthcare company Calico, which use IT to address medical problems, defined themselves from the outset as operating in a world of large numbers. Their research methods are completely different from those of traditional drugmakers. Today, major pharmaceutical companies are working with technology companies, trying to use new methods to solve problems that old methods could not. One could say that awareness of scale determines the business model and the company's success or failure. The difficulties facing Yahoo and similar companies, or large banks, are not because Yahoo employees work less hard than Google's or bank staff are less intelligent than Alibaba's. The difference lies in their thinking. Yahoo and the major banks have failed to grasp the essence of computational thinking. #Dedao #WuJun #LettersFromSiliconValleyGoogleMethodology