"Turn" haha smile

1 I am a programmer, interested in web, distributed and data processing.
I regard the web as the most popular, natural cross-platform presentation and visualization technology, as the most appropriate architecture for program collaboration and collaboration, and treat data processing as a matter of basic, skill, cognition, vision, and domain knowledge, so doing data processing will enhance the foundation, skill, cognition, vision and domain knowledge.
These are nonsense and kinda around, but understanding nonsense, to me or, may not be so simple.

2 I think the ability to ascend from the thick product, so for the foundation of things, should maintain a sustained enthusiasm.
That the foundation should include but is far from limited to:
Data structures and algorithms,
Algorithm
Network (TCP/IP, peer-to, HTTP, etc.),
Computer Systems and operating systems (memory management, cache and cache, threading and concurrency, resource and contention, CPU cycles, local principles, etc.).
Think that the above foundation is never proficient, at least at my level of intelligence. For example, you see the Min data structure, very good, you look at the algorithm introduction, better, can look at the Princeton algorithm, you are very good, you read some papers, and you did with Vim, Emacs, vs, Eclipse or sublime knocked, Then you should be able to see the whole of the blog's algorithm articles in the gaps and levels, then you have a bit of skill.

3 The importance of knowledge structure and project experience will affect each other. The same projects, similar roles, because of the different knowledge structure, bring the promotion and summary will be different, this promotion for the next project or more complex project benefits will also vary, for a longer time, the difference is enormous to the value of 10W or 100W.
With ten years of work experience or one thing repeating for ten years, the difference may be here.

4 think there is a "Java-savvy" kind of rhetoric, but the JVM is bullying. The familiarity of class libraries and frameworks comes from engineering and projects, can be familiar with a simple API, or can be used to understand AOP, can be applied very simple can be very complex. If you focus on performance scenarios, you should consciously focus on the JVM. But the JVM is a vast project, so understanding it is a long-term process, so mastery is a difficult thing to do. But many commonly used scenes, such as BufferedReader and FileChannel, mmap can be an I/O choice, which is more suitable, often depends on the virtual machine inside and operating system to do something, only to come out best practice, with, for XSS, XMX, Simple configurations such as Directmemory will also have a deeper understanding. It also helps to see the limitations of virtual machine-based languages, such as HBase's Blockbuffer itself as a design that improves read efficiency, but because of the JVM's heap and GC mechanisms, this design can drag hbase down.
This should also apply to C#&CLR.

5 of people who have worked for many years will think or feel self-mastery in some ways, some of them do, but with a little outfit, hypocrisy, and a great love of cleverness, he will not discuss the problem with you properly. Such a person may exist in friends, classmates, colleagues, bosses, a programmer will not live in a separate container, in addition to the strength to maintain self-confidence and humility and pay more efforts, but also to adhere to some of their own principles and not by some not enough positive energy corporate culture, Circle culture influence, do their own, In order to better deal with the technology, can give the employer a better sense of the air and real value.
The world is not lack of spray, and some of the jet itself is very powerful, even more than their own, this is the way they live, emotional control is not what skills, you understand the more thoroughly, you can see what is the jet, what is disciplining.

6 on the theory of technical utility, there is a view that the Stars and Stripes should not be known before the United States. For example, not every company, every programmer will come into contact with big data, but human in the absence of flying ancient, has been through the totem and mural to open the exploration and interpretation of civilization, this thing is actually the power of faith, said vulgar is the interest, and then vulgar Point, is the opportunity is always left to prepare people, Willing to not ready to look at the attitude and understanding, ready to be able to use the future, see opportunities, strength and luck is very important, but luck you can not control.

7 Most of human knowledge comes from indirect channels, such as reading books, watching the media, hearsay, leaving school, but not leaving the medium of knowledge. A good source of knowledge is more important than a good way of learning, which is often the main difference between a great college and a bad one. Some teachers in the micro-blog in the growth stage began to teach social computing, if you do not have this knowledge channel, should not recognize the ETL, data mining and inverted index meaning, you may have been skilled in modifying, adding, compiling lucene.
However, the good knowledge, leave the practice, will not be sublimated, think a good programmer, should pay attention to the code and implementation, but should abandon the simple copy principle, although copy most of the work is enough to deal with. Privately, a key code to write a dozen or so silly, quite stupid, but if each writing has a thinking and understanding in the inside, have chosen to do so, and then look at design patterns and reconstruction, oh, really so, with the way of fennel beans when the fool is not so unbearable thing, at least understand what is wrong, what is the difference, It also creates a scene of ascension beyond your work.

8 Many people hold this view, the Chinese, the technology is poor, but also can not write Hadoop, make money the most important, but they feel very bad.
In fact, will write Hadoop and make money as a standard, this view itself is too limited, people in this life, money is of course important, write can use the Hadoop or write mini operating system or mini virtual machine, do not have to take it too seriously, read the source and the realization of their own to create the understanding of the depth of the sky underground, This yugong Yishan approach is more about forcing you to learn more, more accurately and profoundly. Most of the Hadoop experts, in fact, are only looking at the source code, it has been able to out of the book, in fact, many companies in practice, but also just run some very simple, very mature mining algorithm, Iqiyi engineer to do linear regression is also practical, most companies are still processing logs, sometimes think of a very good algorithm, has not been replaced by the SSD important, so there is no environment, without that condition, do not study it? No, I don't think so.
About making money, I think, as long as the interest, can do things well, the money will come. Income does not go, often do things are not high income in that interval, writing procedures difficult to rich. To be honest, spending on the periphery may be several times your salary, needless to say, outside income. Nightclub want to play high point not open two bottles Louis 13, may also be Miss scold pseudo local tyrants, you do not send your favorite red, blue, green card bouquet, you are embarrassed and her back to the community or hotel to explore genetic algorithm.
Then with the periphery to explore the genetic algorithm of the king, Ligong income is not a program can be compared.
So do not deliberately write programs this matter and earn money hang up, limit your ability to make money.

Original link: http://www.cnblogs.com/foreach-break/p/be_a_real_programmer.html

"Turn" haha smile