The key to solving this mystery is to reframe the question. Why don’t smart kids make themselves popular? If they’re really smart, why can’t they find the secret to being popular? They perform so well on standardized tests, why can’t they achieve great success in this area too?
Although “nerds” suffer from being unpopular, I don’t think most people would be willing to give up being “smart” just to relieve this pain. For them, mediocre intelligence is unbearable. However, for other children, the situation is different - most people would accept this trade-off. For many people, this is actually an opportunity to move up to the next level. Even among students ranked in the top 20% intellectually (I’m assuming here that intelligence can be measured, as people seemed to believe at that time), who wouldn’t trade 30 points for others’ friendship and admiration?
The common ground between hackers and painters is that they are both creators. Like composers, architects, and writers, hackers and painters are trying to create excellent works. They are not essentially doing research, although they may discover some new techniques during the creative process (which would be even better).
The way to create beautiful things is often not to start from scratch, but to make small adjustments based on existing achievements, or to combine existing ideas in relatively new ways. This type of work is difficult to express in research papers.
What hackers really want to do is design beautiful software, and evaluating this kind of work is very difficult. You need to have a good sense of design yourself to evaluate whether others’ designs are good. But there is no correlation, or even a negative correlation, between thinking you have a “good sense of design” and actually having it. What can be evaluated is often very boring and doesn’t bring surprises to the world.
The only effective external evaluation is time.
I don’t know any hackers who like programming in statically typed languages. What we need is a language that can be scribbled and erased at will. We don’t want to sit upright, carefully placing a teacup full of various variable types on our knees, carefully choosing words to ensure variable type matching in order to appear polite and thoughtful when talking to the meticulous compiler.
Single-person creation and multi-person collaboration are different. Single-person creation is like scribbling with a pencil, which is suitable for loosely typed languages and allows for constant refactoring. But when it comes to multi-person collaboration, everyone needs to discuss protocol interfaces, types, and other contracts, with the purpose of improving overall efficiency and quality.
Universities and laboratories force hackers to become scientists, and companies force hackers to become engineers.
I found that in their view, the “hacker’s” job is to implement some function with software, not to design software. There, programmers are treated as technicians, whose responsibility is to implement the product manager’s “vision” (if that’s how the word is used).
One of the least known advantages of the open source movement is that it makes learning programming easier.
Programs are written for people to read, with the added benefit of being able to run on machines.
When you look at old photos and see how you used to look, don’t you feel embarrassed? Did I really dress like that? Yes, you’re not mistaken, you did dress like that. When we dress, we have no idea how silly we look, thinking we’re fashionable.
It’s no accident that Silicon Valley appeared in the United States, not in France, Germany, Britain, or Japan. People in those latter countries always act step by step.
Maintaining tolerance for appropriate disobedience doesn’t cause much harm, but is very beneficial to America’s national advantage. It enables America to attract not only smart people, but also those who are very conceited.
Are civil liberties really the cause of national prosperity, not the result? I think so. In my view, a society where people have freedom of speech and action is most likely to adopt optimal solutions, rather than solutions proposed by the most powerful people. Authoritarian countries become corrupt countries, corrupt countries become poor countries, and poor countries become weak countries. There is a Laffer curve in economics that suggests that as tax rates rise, tax revenue first increases and then decreases. I think the same is true of government power - as restrictions on civil liberties continue to rise, government power first increases and then decreases. At least for now, our government is likely stupid enough to actually implement this experiment and personally verify this view. But tax rates can be raised and then lowered, while once this experiment makes a big mistake, it’s too late to regret, because once a totalitarian system is formed, it’s very difficult to abolish.
Now, upgrades no longer have a major impact on users. Over time, software becomes more powerful. This requires developers to put in some effort. They must correctly design software so that it can be upgraded smoothly without confusing users. This is the new problem facing internet software, but there are solutions. All users use the same version of internet software, and bugs are corrected immediately upon discovery. So it should have far fewer bugs than desktop software. At Viaweb, I remember that the most unresolved bugs at any one time were only ten in total, and most problems were solved as soon as they were discovered and wouldn’t be left behind. This is much better than the PC internet era, which has already transitioned from client applications to web applications. The mobile internet era still feels like it’s in the app world.
There is a subtype of compound bugs: two bugs that compensate for each other, like “negative times negative equals positive,” and the software actually runs normally. This type of bug might be the hardest to discover. When you fix one of the bugs, the other bug is exposed. At this point, you feel like you just fixed the wrong thing, because that was the last place you modified, and you suspect you did something wrong there, but you’re actually right.
In fact, because there are few bugs, you only encounter them after some complex process, so advanced users often feel proud when they discover bugs. When they call customer service, they mostly speak in a victorious tone rather than an angry one, as if they’ve scored against us.
The more people there are, the longer it takes to meet and discuss how various parts work together, the more unforeseeable mutual influences there are, and the more bugs are generated. Fortunately, the reverse of this process also holds true: the fewer people there are, the more exponentially efficient software development becomes.
Piracy is not an “advantage,” but it’s also a problem. A certain amount of piracy is beneficial to software companies. No matter how much you price your software, some users will never buy it. If such users use pirated software, you don’t lose anything. In fact, you actually gain, because your software now has one more user, which gives it greater market influence, and this user might buy your software after graduation.
Viaweb initially targeted individuals and small businesses. I think this is the general rule for internet software. These customers have flexible decision-making and need low-cost new technologies, so they are more willing to try new things.
When the web was first born, it wasn’t intended to be used as an application interface - it just happened to meet the need. For a considerable number of users, being able to use the software itself by opening a browser is attractive enough, and the ugliness and inconvenience of the user interface is not a serious problem.
It’s much easier for a few hackers to figure out how to rent offices or hire salespeople than for those companies (whether large or small) to figure out how to write software correctly.
If you want to make a million dollars, you have to endure the equivalent of a million dollars of pain.
If starting a business were that easy, everyone would do it.
The advantage of getting rich by creating valuable things is not only that it’s legal (many other methods are now illegal), but also that it’s simpler. You just need to make something people need.
Wealth is your goal, not money. There’s no need to distinguish between money and wealth here. Money is the most convenient way to exchange wealth. The essence of money is the ability to mobilize resources. In a normal market environment, having money means you have wealth.
The largest remaining group of craftsmen is programmers.
Work is working with many people in an organization to make something people need.
CEOs, movie stars, fund managers, and athletes all have a sword hanging over their heads that could fall at any time. Once they mess up, they’re finished.
Large companies are like giant Roman warships, with a thousand rowers working together to push them forward. But two factors prevent them from going fast. One factor is that each rower can’t see how their harder rowing makes a difference; the other factor is that a team of a thousand people greatly averages out any individual effort.
Ideally, you form a team with others who are willing to work harder to seek higher returns (compared to working for large companies). Because startup teams are often formed spontaneously, many ambitious founders already know each other (or at least have heard of each other), so their assessment of each other’s contributions is more accurate than in general small groups. Startups are not just teams of ten people, but teams of ten like-minded people.
The larger the team, the closer each person’s contribution is to the overall average.
Startups are like guerrillas, preferring to choose hard-to-survive deep mountains and old forests as their base, where government regular forces can’t pursue them.
After really starting a business, your competitors determine how hard you have to work, and they all make the same decision: we can work as hard as you can.
The effort and return of starting a business are proportional overall, but not proportional individually.
The mosquito’s only defense is that as a species, they are extremely numerous, but as individuals, they are extremely difficult to survive. Startups are like mosquitoes, often having only two outcomes: either win everything or disappear completely. You usually don’t know which outcome you’ll be until the last moment.
The most basic principle you must always remember is to create things people need, which is to create wealth.
The consequence of slow work is not only delayed technological innovation, but it may also stifle technological innovation. Only with the incentive of quickly obtaining huge benefits will you challenge those difficult problems; otherwise, you won’t want to touch them at all.
The Cold War, World War II, and most wars in modern times illustrate this principle. We should encourage everyone to start businesses. As long as we know how to store wealth among the people, the country will become strong. Let the nerds keep their hard-earned money, and you’ll be invincible in the world.
Different skills lead to different incomes, which is the main cause of wealth inequality.
The value of a person’s work is not determined by the government, but by the market.
When we say some jobs are overpaid and others are underpaid, what do we really mean? In a free competitive market economy, prices are determined by buyers’ demand. If people like baseball more than poetry, then baseball players’ income will be higher than poets’ income. To say that a certain job is underpaid is equivalent to saying that people’s demands are incorrect. Of course, people do demand incorrect things. What’s so strange about that? Don’t you think it’s stranger to claim that a certain job is underpaid? If you think that due to people’s incorrect demands, certain jobs are underpaid and unfair, then this world will definitely make you feel very regretful. People just like watching TV reality shows instead of Shakespeare, and people just like eating corn dogs instead of boiled vegetables. Isn’t this unfair? If you think it’s unfair, then you’re being as unreasonable as saying blue is the most beautiful color or saying square is round. It’s like the frequent discussions on Weibo about the income of celebrities and scientists. Celebrities’ income is far greater than scientists’, but what can be done about it? Demand and market determine everything.
When we discuss “unfair income distribution,” we also need to ask where income comes from and who actually produces the wealth behind income. If income is completely distributed according to the amount of wealth created by individuals, then the result may be unequal, but it’s hard to say it’s unfair. Fairness and equality are completely different, and egalitarianism breeds laziness.
The second reason many people are dissatisfied with wealth inequality is that for most of human history, the most common way to accumulate wealth was actually theft. In nomadic societies, it was stealing others’ livestock; in agricultural societies, it was taxation (in peacetime) and direct plunder (in wartime).
What rich people do in their daily lives is also similar to ordinary people. A leisurely life of idleness has long become a rare situation. Today, there are indeed many people who are very rich and don’t need to work anymore. The reason they still work is not because they feel social pressure, but because idleness makes people feel lonely and depressed.
While technology increases income inequality, it reduces most other gaps.
In general, what you want to avoid is absolute poverty, not relative poverty. Promote individual wealth creation, eliminate absolute poverty, allow social wealth inequality, but not too much. Common prosperity is not equal prosperity.