People who don't like to innovate, or who have not been exposed to original work, often have the illusion that whether it is invention, classic divine comedy or heirloom poetry, it is the product of a flash of genius.

Mozart's biographer Otto Young once exposed a lie: as early as the beginning of the 19th century, there were many "chicken soup texts" that rendered Mozart's creative process: it is said that the most awesome works of his life were created in the dead of night, in a beautiful environment, and in a state of unity between nature and man.

But in fact, this kind of "chicken soup article" argument is pure nonsense and an outright lie. Otto Young has long proven that Mozart was not such a person.

Even if it is as strong as Mozart, composing music requires slowly studying music theory and repeatedly revising the tuning audition.

Even if he is as strong as Li Bai, in addition to the three "Qingping Tunes", which are known as the three "Qingping Tunes" that were achieved overnight, most of his other poems also have to check the rhyme, pick out the words, and Jia Dao's sentence "The monk pushes the moon and goes down the door", in order to repeatedly "scrutinize" the words one by one.

Overemphasizing the "inspiration" required for creation, while selectively ignoring the "hard work" required for creation, is nothing more than chicken soup and jokers shamelessly catering to the public and satisfying the public's self-psychological protection for the sake of their own articles.

Because the mediocre masses need to find an excuse for their incompetence in innovation. They are willing to believe: Look, creating such a thing is something that only genius can do. We didn't do it just because we didn't have that talent, not because we didn't work hard enough.

In the modern frontier enterprises -- as long as they are not the kind of enterprises that can only copy other people's products, as long as they have some real independent optimization and innovation -- in their R&D departments and software departments, there are engineers and programmers who rely on hard work and hard work to achieve creation. They don't have the kind of genius that the layman thinks, all they have is hard work and a right direction.

Whether it is Zhu Haibo or Du Junqiang, they are just an insignificant wave among the millions of original code farmers in China in the new century, they have no genius, but they also know hard work.

As for whether they have grasped the right direction of effort, they dare not say before that. It is normal for code farmers who have just worked for two years to feel confused.

But after meeting Gu Mojie, they dared to speak. Because they know that what they are looking for is there in Mr. Gu.

……

Less than half an hour after Gu Mojie's phone call, the takeaway was delivered, and he beckoned everyone to come over and eat together, and he also picked a rice bowl first, and ate it directly, and he couldn't see the boss's shelf at all.

The skeptical programmers sat in a circle, ate a few bites, and no one was embarrassed to ask first. Until Gu Mojie swallowed a mouthful of shredded green pepper meat, waved his chopsticks and asked, "I said that I took the opportunity to have a regular meeting at lunchtime, so you don't have anything to ask?" I am only responsible for answering questions in today's meeting, and there are no hard topics. ”

Hearing Gu Mojie say this, Zhu Haibo was the first to hold back and asked, "Mr. Gu, I want to know, how many people are in our company's testing team?" What qualifications are they? To be honest, I have never seen such a strong testing capability in Sohu for two years, and I can actually test so many problems for a code that has not yet run through the basic functions. ”

You may wonder why Zhu Haibo cares so much about the company's testing strength. This problem can be illustrated by an analogy.

The programmer is like a patient waiting to be treated, the programmer is the doctor who prescribes and dispenses the medication, and the tester is the person who observes the patient's symptoms, gives feedback on the patient's symptoms, and the programmer's "patient's drug reaction" after taking the drug.

The worst tester can only tell the "programmer doctor" that the program "hurts". A better tester can tell the "programmer doctor" what "is sick" with the program. Therefore, any software company needs the cooperation of two adversarial departments, the software department and the testing department, in order to produce a qualified software, just like the prosecution and defense need to argue in court, so that the truth can become clearer and clearer.

Whether a company's testers are awesome or not will have a great impact on the growth of programmers' abilities.

For example, if a mobile phone solution is faced with a fault with the symptom of "the mobile phone screen does not light up after flipping", the tester will only submit a bug according to the above text in a timely manner, and then the programmer needs to spend his old nose to find out what the problem is. A lot of energy was spent, but not much progress was made.

Experienced and discerning testers, when they encounter the same bugs, will not be satisfied with the above simple conclusions. He will use more control group experiments to investigate the root cause of the problem, and it is likely that he will finally conclude the deep problem of "abnormal communication code of Hall device sensor on the clamshell". By cooperating with a powerful tester, programmers can save a lot of detours and make their experience and coding skills grow rapidly.

The strength of the company's testing resources is closely related to its own vital interests and the speed of progress in its future work ability.

Gu Mojie listened to this question, smiled indifferently, and threw out the answer that had been prepared a long time ago: "The company must have a strong testing team, you can rest assured." However, there are some personnel who are still outsourced and cannot be introduced to you. ”

Zhu Haibo listened to this overly official answer, and was even more itchy, and asked further: "Mr. Gu, I absolutely believe what you say, and I don't have any doubts about the strength of the company's testing team." But I really haven't seen such a strong test in China, can you tell me some technical details?

For example, for the part of the code I wrote yesterday, I saw that my colleague in the test raised a bug of 'there is a memory leak in the word frequency feedback data scraping link', and it was accurate to a certain piece of code, which is simply amazing. The basic functions have not been run, how did the test get to this point? ”

The more complete a software is, the easier it is to accurately detect bugs, and the more primitive and basic the software, the more difficult it is to accurately detect problems. This is true all over the world.

Just like a mobile phone, if you can't even turn on the basics, you can only detect one bug: it can't turn on. Even from the code level, the phone has problems with camera data transmission and touchscreen algorithms...... It's not even my turn to be measured.

But this kind of thing that seems impossible to ordinary people, Gu Mojie just did it.

The part of the code that Zhu Haibo wrote yesterday is equivalent to a mobile phone that can't be turned on. And Gu Mojie is equivalent to saying "if this mobile phone is turned on, what other bugs will appear" when the phone is not turned on.

If the testers of ordinary Internet companies belong to "seeing a patient and saying where the patient hurts"; The top testing experts in China can "accurately tell the cause when they see the patient"; Then Gu Mojie's test team, yesterday's performance can be compared to "even if you see a corpse, you can directly tell the cause of death of the corpse without an autopsy".

Forensic doctors are more competent than doctors who treat living people. A person who tells the cause of death without even having to perform an autopsy is naturally stronger than a forensic doctor. This kind of testing ability can already be said to break through the sky. It's no longer just a vassal of programmers, but a mentor and friend of programmers.

In the midst of everyone's curiosity, Gu Mojie put down the rice bowl, held a chopstick, and pointed out the country: "That's my own software for the company to develop, wrote some code-level automatic test plug-ins. Therefore, any structure is incomplete, pointer calls are undefined, and memory leaks are not self-consistent...... and other low-level errors, which can be detected. Even if the code is incomplete, you can find problems. ”

As soon as Gu Mojie said that testing software or testing plug-ins, Zhu Haibo, Du Junqiang and others understood a little, and what followed was a sense of awe.

The test plug-in is a kind of software that reduces the repetitive work of testers and automatically completes some simple and repetitive test work through functions similar to "plug-ins".

For example, writing a script with a button wizard and pressing each key combination on a software to see if there is any problem with the function is the easiest test plugin.

If there are some low-level errors in the target software, they can be found with the automated testing of the test plugin. However, the functions that can be achieved by test plug-ins are generally relatively simple, and can only be used for streamlined tests or stress tests. If you want to use a test plugin to find complex problems, it is not unusual, unless the person who wrote the plugin is also a master of the program.

In China in 2003, I had never heard of any company's test plug-in that could do so well, but the test plug-in written by Gu Mojie did it.

……

Gu Mojie answered a few superficial questions from Zhu Haibo and Du Junqiang, and gradually completely calmed down the programmers.

Only Gu Mojie himself is clear in his heart: where does he really have such awesome skills, and can know more than programmers who have studied search engine code for several years? The reason why he can give so many guiding opinions is completely because of the VS2050 awesome thing that Hatsune Musume gave him.

As we all know, the more advanced a piece of software is, the lower the requirements for the user's software operation level.

THIS REASON IS EASY TO UNDERSTAND, JUST LIKE EVERY TIME AUTOCAD IS UPDATED, IT CAN REDUCE THE TIME SPENT BY DRAWING ENGINEERS IN THE DRAWING PROCESS BY 5%; PS Every time a version is updated, it can also make the artists of the P picture faster (the time to conceive the composition in the brain is not counted, here only refers to the operation time of the drawing software)

VS As a tool for programmers to write code, the efficiency between the 2002 version and the 2050 version is also very different. In the future, thanks to big data, VS can count the mistakes and bugs made by all programmers in the world, summarize some common diseases that programmers are prone to, and then integrate these common bugs into the VS2050 automatic test plugin.

So programmers in 2050 will not only write much more readable code than they did in 2002, but will hardly even make low-level errors other than logical errors. With every decade or so of the era, the average coding efficiency of programmers per unit of time has doubled, and this law has not changed from ancient times to the present.

Now that Gu Mojie has opened a company, the client part of VS2050 naturally cannot be directly used by the company's programmers. But because of encryption on the server side, only Gu Mojie can access it, so no matter how big his hands and feet are on the server side, no one will find his golden thighs.

In this way, this plug-in is logical and logical. When programmers write code, many errors cannot be directly corrected automatically, but as long as they are uploaded to the company's code server, and the server runs the automatic test function, many problems can be found, and then a layer of skin can be batched, and the bugs can be fed back in the name of a fictitious "company test team".

Of course, when Hatsune Musume traveled back to the original time and space, because the cloud BUG database existed in 2050, Hatsune Musume could not call the database of the later generations, so she could only start from scratch after returning to the original time and space, and collect and enrich herself through various channels on the Internet.

Thanks to this, Gu Mojie borrowed VS2050 to complete the automatic test, so he did not behave too rebelliously and did not exceed the imagination limit of his contemporaries. Otherwise, he would need to hire at least half of the number of programmers.

……

After everyone was suppressed by Gu Mojie, as the explanation became simpler, their hearts became clearer and brighter, and they also made up their minds about whether they wanted to work under Gu Mojie for a long time.

In Internet companies, programmers who receive the same salary often have some who can leave work on time, while others work overtime every day. Outsiders may think: Is it because the boss is unfair and gives some programmers a lot of tasks, while others who don't work overtime are related users?

In fact, this is a big misconception, and among all the innovative industries, Internet companies are the easiest to allocate work according to the amount of tasks.

The reason why some programmers work overtime is entirely because they write too many initial code bugs, so the follow-up repeated correction workload is too large, and the tinkering leads to overtime every day.

And if you really have the ability to be like those program bulls, you have written it over and over again, and there are only a few small bugs, so you don't need to tear it down and start over on a large scale. It is completely possible to do without overtime, or even leave work early.

There are only two possibilities for you to work overtime:

First, your original code is too badly written, and the amount of subsequent modifications is huge.

Second, the testers you work with are too bad to pinpoint the problem for you, so every bug you have needs to repeatedly troubleshoot the core crux of the problem, resulting in you needing to work overtime.

A layman doesn't understand this truth, how can a programmer not understand this truth?

Gu Mojie's test team, which has not shown up so far, and the few amazing test plug-ins written by Gu Mojie himself, can actually detect the bugs in the code that the main function can't run, what kind of ability is this?

And there is another question: if there are really testers in China who have achieved this level, who will be willing to be only one tester? It is important to know that domestic software companies do not attach importance to testing, and the salary of testers is only a fraction of that of programmers.

So, even if they go to Baidu, Ahri, and Tengxun to work, can they make themselves grow up quickly like they work here in Gu Mojie? Do you benefit from your ability to code every day?

Apparently not.

Gu Mojie borrowed Hatsune Niang's VS2050 in exchange for such an effect: If you want to work with others for a high salary for a while, maybe I can't keep you. But if you are an aspiring programmer, not satisfied with just making money, but also want to be obsessed with writing code, you have a little ambition to improve the world, and you want to quickly improve your coding skills. Then they will be attracted by Gu Mojie, like a compass meets a magnet, and they will be conquered by Gu Mojie.

"Mr. Gu! I apologize to you. Zhu Haibo's legs were weak, he sat on the sofa, drenched in cold sweat, as if he was seriously ill, "I doubted you yesterday, and I thought that you were unreliable here and couldn't learn real skills." Now I know that I really have no eyes! As long as you don't fire me in the future, I will be mixed with you for the rest of my life, and I definitely don't want to change jobs! ”

Gu Mojie smiled slightly, looked around the audience, and several other programmers were also completely conquered by his performance, with a posture of wanting to cut off the head of the chicken and burn the yellow paper to show their loyalty.

Some also said that they knew a few colleagues who originally went to Baidu and Ahri, and now they don't want to change jobs, they will definitely work hard to help Gu Mojie poach people, and if they can't dig it, they will publicize Gu Mojie's awesomeness to those former colleagues, so that they can see it with their own eyes......

After listening to the last statement of expressing loyalty and helping to poach, Gu Mojie finally breathed a sigh of relief.

What is most important in the 21st century? Talent!

This domineering atmosphere is really not easy to leak.

……

(PS: I can only bow to kitsch, and I wrote a chapter that is biased towards the reading preferences of programmers for two days, and I lost more than 50 favorites.) It seems that the social class of programmers really has little potential for consumption.

I'm not going to die, I'm going to start going to the college campus chapter tomorrow. This chapter is nearly 5,000 words, which can be regarded as a conclusion. I didn't recommend Guoben this week.,It's still so much updated.,It's better to reduce some words for the New Year.,No one is chasing anyway.。 ）