Written in the programmer transferred to PhD year four

06 After graduation in order to feed the family did not continue to read the doctor, transferred to the company to do software development, 2012 think life can go to the Chinese Academy of Sciences to study.

The program has a crazy hobby, the longest programming time is 24*3 hours, in addition to the toilet, lunch without any pause. Not computer science, just hobbies, hope to write this blog to encourage people who like to engage in program development, and love programming.

Do not feel inferior because you are not a computer professional or a software development professional student. Many factors will affect people's psychology and decision-making, whenever there is interest, people can overcome the environmental factors, professional just a point in time you can be subjective or objective to be asked to do one thing, and do not spend much effort. Real computer enthusiasts should be such a person, something good willing to expand their knowledge of the computer, to read to find some of their own unclear or do not know the technology or theory. If you are determined to do something, you can become an expert or professional in this field for 5 years. Do software is my hobby, learning is the first way to do a good job in software, interest is to be able to adhere to the first momentum of the long-term.

What a real programmer. is not limited to writing code, but is the transformation of domain knowledge. If you stay in the details of the code, you can only become a skilled code worker, to earn the same amount of money, but there will not be much development. The real programmer is a kind of belief, can transform the demand, can abstract the demand as the domain model, uses the abstraction ability high and low to measure the programmer's difference, the programmer oneself at different stage difference is most effective. Abstract ability is the basis of the establishment of procedures, any discrete mathematics, combinatorial mathematics, or both the basis of the set theory, are through the most essential abstraction to solve the problem. The process of solving complex problems through logical combination is the process of abstracting concrete problems as models. Whenever a tricky problem can be mapped to some sort of mathematical problem, that is the embodiment of the programmer's ability. Except for this, none of it.

How to train your own computer programming skills. Kung Fu is not in the training of programming, but in the perception of problems. The so-called perception is the reflection of the problem as a mathematical problem. The University stage programming is the propositional programming, gives the topic, uses some conventional thought to solve. The doctoral stage is seldom programmed, and the main focus is on how to understand the problem, transform the complex problem into some mathematical proposition, and solve the proposition of the estimated solution. University stage desperately reading is about the library and model books, from Peking University Press Win32 programming, to Tsinghua University press data structure, and to Martin Fuller Enterprise-level development, and then to Gang of four model design, or Posa series, these are my special want to understand, Maybe a lot of people think that this can be complacent, but this is just the beginning, to master these, can only be said to be a slightly advanced programmer.

What is the best programmer. Programmers who see the commonalities and features of the problem. The development of modern scientific research has turned to positivism, that is, the correctness of the reaction theory through experimental results. If the programmer is the same, but the programmer can not pass the experiment, but through the project to verify his theory. The higher the value of the better programmer, the higher the cost of the failed project. Patterns can only be given in the form of coarse-pan generality, the real essence is precisely defined by mathematical principles. Different areas of knowledge determine the characteristics. This attribute can be different from the semantics that can be different form.

The different effects of PhD-phase learning and undergraduate-level learning on programmers ' abilities. Formalization and solution are the eternal themes. I personally think that the work of the PhD students in the major of computer science and mathematics is a kind of effort to formalize the work, and the need of solving methods on this formalized basis. is not constrained by the algorithm, but a deeper understanding of the algorithm. The best example is the development of artificial intelligence, the luger of artificial intelligence, the name of the reader is the programmer, in deductive reasoning to pursue formal proof and recursive solution, in the inductive reasoning aspect of the pursuit of symbolic representation, connection representation, resulting in the development of symbolic reasoning, classification and clustering methods.

More attention in undergraduate and working stages on how to use or master a programming language, programming more time; But at the doctor's stage, I felt most about designing a language I needed and thinking more.

To the young people around, do not think that reading useless, but the level of abstraction is too low will hinder the depth of thinking.

from:http://blog.csdn.net/jx1228/article/details/50349839