Programmers are not bricklayers, they are writers English text: Your Developers Aren \u0026 rsquo; t Bricklayers, They \u0026 rsquo; re Writers If you have 10 programmers, the best that could at least five times better than the worst. This is absolutely not nonsense. We define \u0026 ldquo; bug less work better \u0026 rdquo ;: faster generated code more readable, logical and maintainability. Programmers are not bricklayers, but they tend to be treated as a bricklayer. (I'm not saying discrimination against these occupations) \u0026 ldquo; Why do I need a senior programmer, to know the same salary I could hire two of the primary? \u0026 Rdquo; \u0026 ldquo; This function requires a programmer to do three months, it only need to add two, you can get in a month. \u0026 Rdquo; why the above idea is ridiculous? Because we do not have a simple and effective way to measure programmer productivity. Once encountered something we can not measure, we will ignore it. Hello I ask: you are willing to let the two newcomers to take care of your baby, repair your car, you do a lumbar puncture, or would you rather look for a veteran? Research shows that the best programmer most productive programmers than 28 times the worst high. However, the best programmers who use these costs would not have so much, so they are the most cost-effective software field \u0026 ldquo; \u0026 rdquo ;. Specials ROBERT GLASS, 'FACTS AND FALLACIES OF SOFTWARE ENGINEERING' If you have to compare, then the programmer actually more like a writer. Some writers can write something millions to sell, and some writers write things bored and finally only to light a fire! However, they have produced a book, therefore, they are writers. But unless you go read their books, otherwise you would not know the difference between the two of them. Program managers had long recognized the difference between a good programmer and programmer productivity with both of the other days of the capsule. However, the results of the actual measured data still let us all by surprise. In the study, Sackman, Erickson and Grant wanted to measure the performance of a team of experienced programmers. The results showed that the best and worst average productivity ratio of about 10: 1, in particular the proportion of the programming speed of surprisingly up to 5: 1! FRED BROOKS, 'THE MYTHICAL MAN-MONTH' Let me tell you a true story. (For the name has been changed.), A software company needed to implement a new module in their logo products. Mr Lousy (Mr. difference) is just empty, so will this task to him, so he immediately started! After months of tinkering, Mr. difference finally completed. QA team found that there are a lot of bug and contradictions, and will report back to Mr. difference. According to the report, Mr. poor and spent two weeks time to fix the bug, and finally gives a new version! Those damn annoying bug really racked his brains Mr. poor. Test and then repair, so the la corda doro wigs cycle a couple of times. Users are already looking forward to this new module. Sales staff also signed a number of new customers. This module to do it, and be integrated into the next version imagined the pressure of the process. But fortunately, a success! Champagne bar! Yeah, no, the new module is still full of bug. People's eyes are always sharp, always particularly good at customer ago to find some undiscovered bug. So they contact technical support. Technical support team to look for what went wrong, the customer does not know how to operate the function, or they themselves made a mistake, or whether this is just a bug, a reproducible bug. \u0026 Hellip; \u0026 hellip; and technical support team submitted a bug report. So, Mr. difference tragedy, \u0026 mdash; \u0026 mdash; I mean, even if he is engaged in another project, this time can only drop all the work at hand to solve these troubles. (This does not relate to maintenance, logic and documentation issues codes \u0026 mdash; \u0026 mdash; because in the future there will certainly be other people to improve the code.) But, alas, how should we say \u0026 hellip; \u0026 hellip; seems to have some bug beyond Mr. poor capacity, he simply could not repair. Moreover, emerging new bug, nobody really knows they are new, or whether there before, but that is not to be found only in. Support bored. While sales more annoyed, because new customers have said that this module so do not fly, they have to cancel the contract! Finally, the boss had to let Mr Rockstar (good sir) to look at the code. Good sir did not want to clean up the mess Mr. poor, many of the structures in his eyes is meaningless. He suggested rewriting the code, it is expected to need about a month's time. Then he started up, only one more than the mew mew wigs original estimate of a few days to complete the module (he was good sir, not Mr. Perfect). In addition to QA team found some small mistakes, everything will work as expected. QA team silently worry: If all programmers are like him, they have no useful here. Mr. difference think this is good sir arrogantly laughed at him, but his views on good sir terms is irrelevant. Improved module off quickly released. Everyone is happy, because finally working. Hooray! Now we can open the champagne to celebrate the time! But this time, everyone seems to have forgotten the good sir only a month or so to get Mr. poor success with seven or eight months is also not handle the task. Huge difference between the two levels of developer productivity is evident \u0026 mdash; \u0026 mdash; they perform identical tasks. By writing a leaner but more functional code, bug by writing less code easier to maintain, a good developer black butler wigs can reduce QA staff, colleagues and managers work pressure, improve the productivity of everyone around. This is why the research results of these data, such as 28-fold productivity is possible, even likely also reported low, if code geass wigs you the big picture, then. PHIL HAACK, '10 DEVELOPERS FOR THE PRICE OF ONE' Well, the worst thing in this story that may occur, what is it? Good sir finally noticed him than Mr. poor much more efficient: He can easily identify malicious code. But nevertheless, he was sure his salary and Mr. difference almost. He expressed dissatisfaction, even resolutely leave. So your team lost a pillar of strength and efficient. Even if he does not leave, when faced with due to the release / project delays caused by the entire team of overtime, what would he think? Departure is inevitable, just a matter of time only. And, if you really really bad luck, to mention another good person to replace Mr. position, but unfortunately is another difference between Mr., then I can only silently as you point the candle. So why do we always used to ignore the facts? Because much easier to ignore than to correct the problem \u0026 mdash; \u0026 mdash; at least to some extent, that's true. And no one is willing to do informant colleagues villain, as he lost the job of reasons: because of the Diocese Mr. they might want to support poor, starving children under; perhaps he had just bought a new house, every month repayment \u0026 mdash; \u0026 mdash; most importantly, it is hard to measure the productivity of developers, unless you let them do the same task to do both contrast, the same as the above story. So I've been thinking about this question: how to measure the efficiency of the developer. I am very jealous in sales because their pay is angel sanctuary wigs based on performance come. I have some ideas (not mature) lets you select the essence to its dregs. I also have some ideas about how to identify, attract and retain good developers. Why is my identity, and I want to tell you the truth I worked as a full-time developer for ten years. Early in 1989, I made my first commercial product (game). Although Cosplay Costumes it did not give me a lot of money, but I feel really super good (I was only 16 years old). A few years later, I sold one game idea, and eventually released on the Nintendo DS (and other formats) on! From the perspective of my experience, I can tell you, when you see something you want to come out (including the name) eventually appear in stores, it feels not too cool. In 2008, I was a candidate in a technology-driven company in a non-technical positions. I would like to know the general mode of operation of enterprises, all non-technical processes occur, including sales, including the enterprise. Although, my technical ability is helpful for my job, but is no longer an important part of the work. I do not for a living program (to be exact, was), but often tinker with a variety of new technology in the amateur project. For me, reading a good book, not only exciting, but also be able to relax. In my current position, I often encounter some misunderstanding or lack of development of basic concepts of knowledge. While living in their positions (usually those who CxO), the basics will be a huge impact. Many developers CxO will be deemed to be bricklayers, rigidly calculate their efficiency. But here, I would like to reiterate once again that \u0026 ldquo; \u0026 rdquo ;, writer theory developers are intellectual workers, OK? Translation link: http: //www.codeceo.com/article/programmer-not-bricklayers.html Translation of: Code Rural Network \u0026 ndash; Komine