Why should we be proficient? Ten years of noise and other things

30 years old. In the past two or two years, I often read the first "Jin Se" by Li Shangyin:

Jin Se has fifty strings for no reason, and a string is a Sihua Nian. Zhuang Sheng Xiaomeng fan butterfly; hope the Emperor spring heart to trust the cuckoo. There are tears in the pearl of the sea and the moon. This situation can be recalled, but it was not clear at the time.

This situation can be recalled, but it was not clear at the time. Looking back on the past ten years, it has been a quick start. Over the past ten years, I have experienced many fields: new materials, writing, technology management, enterprise analysis, software development, and so on, all of which have left a deep footprint: materials Science is my original major and successfully obtained a master's degree. Other achievements are not ordinary. I have published novels, published books, hosted projects, and founded a company. In retrospect, there is only one word: impetuous.

In terms of development, it started the mud Program Development in 1997 and started the network wave stage, ASP hehe. Later, it naturally went over to Asp.net, C #. As the project requires, C ++, Java, TCL, python ...... I have used many of them. I was also obsessed with the model. During that time, I bought books from Amazon and paid thousands of books a month. Looking back, there are not many valuable things, but many important things are ignored.

1. First, let's talk about ignoring things.

(1) English.

Ten years ago, at that time, SQL Server was still 7.0. It was only in English and could not find a Chinese character. One day, I want to back up the database and check the options above: backup ...... Restore ....... At that time, the English language was about Level 3. I didn't know it. I guess: Isn't store Storage? Restore ...... Isn't storage a backup? Isn't up? Does backup just pick up the stuff behind it? As a result, I understood the restore as a backup and the backup as a recovery. Result: The data in the last few days is immediately wiped out.

The importance of English cannot be emphasized. First, as far as development is concerned, most technical documents are in English. Reading English can better understand the API. You should try your best to read English documents instead of translated documents. Second, in terms of demand, a dollar is similar to a dollar in the eyes of foreigners in our hearts. There are several times the gap. Good English can better communicate with them and earn these tickets.

Ah, sorry for my dumb English.

(2) Mathematics

It can only be described as "suddenly existed.

Calculus, linear algebra, probability theory/random mathematics, and numerical analysis are the highlights of college mathematics. At the beginning, I thought that I would not learn it seriously if I didn't use it any more.

Random mathematics was the first to be used later. In my thesis, I used basic random mathematical tools combined with a series of simple characteristics to cleverly obtain some of the characteristics of the material. Here I realized the power of mathematical tools. One year later, the simulation of discrete events was also brilliant in random mathematics.

Recently I saw an image recognition project on csdn, and soon I made a prototype. Later on, the recognition rate was not high. To enable the recognition rate, you need to use mathematical tools such as Neural Networks and Hough transformation. Having read more than half of the book "Pattern Classification", I woke up calculus, linear algebra, and numerical analysis that had been sleeping for ten years in my brain. Fainting, what neural network, is not just the extreme value. The University's "Numerical Analysis" changed to a vest.

(3) Persistent

It is better to be good at ten things than just one thing. Looking back at the past ten years, my other major mistake is that there are too many things that I am good. Others only focus on one thing, but I need to put it on multiple things. At the beginning, everyone was on the same starting line. Now, college students have almost entered a rising career period, and those who insist on entrepreneurship have also become well-known writers, while I walk with multiple legs, I am not too far away from the starting point. In military words, it is a secret of sub-troops.

(4) Specifications

Learning norms are the cheapest way to learn, and focus on tools. Ignoring them often costs a lot. In my own example, I have developed a spider for a long time, but I have not read the HTML specification. I recently printed the HTML 1.1 specification and read it again, I found that I had many misunderstandings about HTML, and many bugs originated from these misunderstandings.

Return to my current city from Beijing. net, I only brought one 《.. Net program design specifications, but before that, the specifications have been ignored.

Ii. Try again.

(1) Value

No matter what you do, you must use the word "value" in your mind. Something without value is easy to change, difficult to grasp, and difficult to last. Value issues need to be considered when pursuing technology. We learn technology to realize value, rather than simply technology.

The value of technology is embodied in:

· Solve the problem. For example, implement a function.

· Improve productivity. For example, to implement the same function, the introduction of the framework can shorten the development cycle and improve the quality.

· Provide good products or experiences. For example, Ajax.

People who can effectively solve problems are field experts. People who are able to provide sufficient productivity are technical experts. People who can provide good products or experience are product experts.

You can often ask yourself:

(A) What are the issues to be solved?

The development of technology refers to the deconstruct and reconstruction of life. The development of computer technology successfully deconstruct and reconstructed the print technology, the development of contraception Technology successfully achieved the separation of sex and reproduction.

Can we use technology to separate delicious food from delicious food? Can I have a programmable delicious chip?

This problem is very interesting, but it is too big. It may not be implemented in the short term for 10 years or 20 years.

Change to a smaller one. Take the stock market as an example. The expectation of the stockholders has a great impact on the stock market price. The public opinion of the stockholders can reflect the expectation of the stockholders in some aspects. The Internet is a big public opinion platform. If we can discover the association between public opinions in the stock market and future stock prices, and measure public opinions in the stock market through Internet Public Opinion Monitoring and Data Mining, such a system is of great value.

(B) What methods can be used to increase productivity?

There are several factors that affect software productivity. Human resources are a factor, and the development process is a factor. The technical infrastructure is a factor that requires comprehensive consideration for technical decision-making, instead of blindly learning new technologies and learning new methods.

(C) How to improve user experience?

Sometimes, when we get a new tool, we often forget why we need it. Taking Ajax as an example, AJAX can improve user experience, but it can also damage user experience. For example, I often cannot log on to my Gmail mailbox. I have to clear the history data and clear the JS cache to log on normally. For another example, the China-Pub search function in the past may not be used in 80% cases. In this way, it violates the value it wants to achieve.

By the way, let's take a look at csdn and how it looks like a grocery store.

Technology can only be a media, but it may change rapidly, but its value rarely changes. Just like men and women, it's nothing more than changing the vest.

(2) Competitiveness

Hey, valuable things may not be competitive. For example, water and life are indispensable, but they are super cheap. It's easy to get. When there are more homogeneous things, the competitiveness will come down.

As far as technical personnel are concerned, the competitiveness mainly focuses on:

· Going deep into one thing is better than other people. If you get it done thoroughly, you will naturally be competitive. This path requires patience.

· Subdivided one thing to make a pattern based on satisfying the basic needs of the customer and meeting other customer needs better than other competitors. To take this path, we need to study and grasp the needs.

· Looking forward, before something happens in the future, we will get something to wait there. Let's see the world: give me away. This path requires insights and perseverance, and a lot of luck.

A very interesting discovery has recently been found to be active on the Internet.

There are two types of feelings:

The longer you live, the less you get paid per unit time. Because these jobs are not urgent and highly competitive, publishers are psychologically positioned at the market price.

The shorter the time you take to live, the more you get, because these jobs are urgent and there are not many people to do, or even more people to do, however, there are very few people who can make it in a short time. The faster you do, the more money you get. The publisher locates Based on the psychological price.

What kind of work are you suitable? As technical personnel, we often need to think about our own competitiveness.

(3) let's take a look at Mao Xuan.

Iii. Comments on some technologies (personal position)

(A). Net 3.5/Vs 2008

What's best is that you don't need to write specific get and set for the attribute, which makes the code clean a lot at once.

The extension method is a good gift: it can make our focus more focused on objects. Without it, many helper classes have to be written before, and when it comes to specifics, we should carefully flip these helper classes to waste time and brain cells. It's especially smooth to work with activerecord.

LINQ: How does it look like a vest. No need to use it yet.

Lambda: The type cannot be automatically deduced under C #. It can be used only to save some statements, solve a problem (less write statements), and introduce new problems (more grasp a little syntax). It is useless.

Silverlight: Not interested.

SOA: Not interested.

(B) Oo

DP: I forgot about it. I only remember one thing, that is, the rule mode. After programming for a long time, this is the same thing as water, and it is natural to write code. On the other hand, the longer the principles of OO design are, the deeper the memory, and more experiences.

UML: if it is not a big project, it is useless.

4. Excellent programmers in my mind:

(1) There are 1-2 fields that you are good. For example, Web Front-end development should be good at HTML, CSS, and Js. If it is business-oriented, you must be good at requirement analysis, modeling, and communication.

(2) master a software development process suitable for teams, companies, and customers.

(3) be familiar with related technical infrastructure (basic tools and frameworks)

 

This article from the csdn blog, reproduced please indicate the source: http://blog.csdn.net/HuDon/archive/2009/07/16/4353756.aspx