I want to be Daniel, first team, don't fight alone

I've studied the fact of the ape-in-the-way project, are they really that good? How did they become stronger? (It is possible that they also have a general level).

watermark/2/text/ahr0cdovl2jsb2cuy3nkbi5uzxqvc2f5ywxpdgluzw==/font/5a6l5l2t/fontsize/400/fill/i0jbqkfcma==/ Dissolve/70/gravity/center ">

I've read it recently. Mark Pearl a blog post. The title is "programming, a subset of writing".

I agree with most of the points in the article. Read such as the following paragraphs, really touched me, and made me shocked:

No doubt, just like the skills in life. We want to get better. On the line to continue to exercise. When you're done refactoring the code. What else can you do, a further refactoring code?

We open our hearts and let other important things come into our programming, so. Will and that red to improve our programming ability, and can bring a lot of other inspiration. Unit test, exception handling. Even the same communication skills are very important.

Suppose you neglect these things. Self-talk become an overconfidence, self-willed person.

I often inspire the program ape to become an excellent program ape. and compliance with the practice of religion, and ignoring the usefulness of the key factors, that is, to abandon the use of the principle of learning. I thought it was an excellent program ape, and after careful deliberation, they were merely arrogant procedural apes. However, it is fair to say that among them. Some programs ape is still very good.

I very much agree with one of my former colleagues Russell Politzky views, such as the following:

Those arrogant procedural apes are dogmatic, parochial, unrealistic, and so on.

possess such characteristics. Nothing good in our business will lead to inappropriate, even destructive, design decisions.

These programs ape, in peacetime work exchange. Statements such as the following may often be used:

· All functions must be tested by the unit

· All of the test content, to cover all the code

· All the simulations / stubs, etc., used in your unit test module, must be built into the simulation library .

· All applications are conducted around DDD , no matter what their function

· all applications, when building data access layers, you must use ORM to complete

· application, assuming that it does not work with ORM builds, will be the next worst, and very difficult to maintain

· The best stare is the code itself, when you write the code well enough to include the variable name and method name, and the class name, to reach the known name. Therefore, it is undesirable to add code to any gaze.

· Whatever documents or products you have designed. It's going to be pretty fast, and it's going to be useless very quickly.

· The code is the only document you need

· The code is the only model you need.

· a good top-down designer is impossible to achieve and often fails. Those who uphold such principles sighted, until now. They all agree with the superiority of "evolutionary design", which is called TDD.

· In addition to using object-oriented. You can not write good software, because there is only object-oriented talent to reduce the complexity of

Do you know such programs ape? If so, do you know what their characteristics are? Experience has shown that such thinking is extreme. It is illogical and absurd to fits all the things that are not good or bad.

Reasonable reasoning. Focus on the ability to develop a useful sex. Will make you an excellent program ape. Your experience is intact. Improve your skills, which of course is good. But. We have to think about the budget, the cost, the development cycle.

If we can have this ability, grasp the overall situation. Able to become an excellent program of apes.

Of course, become the de facto excellent program ape Vice, hands-on, multi-practice is the kingly way.

I want to be Daniel, first team, don't fight alone