Unconsciously as well as the Nineth week, this should be my last article of this semester reading notes, think or there is a little bit, similarly, through the Dream break code of the last four chapters of the reading, I have a lot of feelings and experience.
First, read the content
In the eighth chapter of the Whiteboard on the instant sticker, the first mention of "Eat your own dog food", meaning that developers must use their own products. The key to better progress is to improve the software to the extent that it is available to them. At the IETF, Doussotte is involved in the development of the new standard WebDAV, and for many programmers, the standard area is a formidable swamp. But Doussotte says when you put a new piece of the system in, you have to look at it for 5 or ten years and you won't regret it, can you expand it and replace it. The working mechanism of WebDAV is to extend HTTP, and then to talk to peers, and it's important to be able, not the architecture itself. After mentioning the spreadsheet, although clear but huge, Doussotte suggested using stickers, each piece of paper to show roughly the same amount of work, each posted on behalf of a developer one months or two months of working hours, so that the whiteboard on the instant affixed to the term.
In the Nineth chapter of the method, the beginning refers to the quality of the triangle, that is, time, money and characteristics, Chandler software download is quite low, although there are many processes, but also useless. So we talked about how to improve the organization's code, and ultimately the gait competition between improving the organization's staff and how they work. One of the good plans is too valuable, but too difficult to make, you have to work harder to plan, spend more time, more details, and then do a plan! Until there is no plan--and then make a little plan. The project management guru then made the proposition that more plans must be made: unless the developer plans and follows the individual work, the work will be unpredictable. The success of the cafe Hanfstingl in IBM is based on two principles: the plan is mandatory and the plan must conform to reality. The values of the approach were later mentioned: individuals and interactions are higher than processes and tools, working software is higher than detailed documentation, customer collaboration is higher than contract negotiation, and the corresponding change is higher than the compliance plan.
In the tenth chapter of engineers and artists, the "Software" and "engineering" are inseparable, the software reliability, quality control, cost control and scheduling, and so on, software from art to engineering, many programmers are not satisfied with such a leap. The role of engineers is to build bridges between the depths of art and science and to predict the reliability of future software that is invisible. The dual identity problem of programming, for decades has puzzled the vast number of programmers, is engineering or literature? I think it is not contradictory, as a work of art, it must have the art of inspiration and innovation, but can not be absolutely blind development, to have the same as the project of the general planning.
In the 11th chapter through the dog food version of the road, learned the Osaf "Dog Food Edition Calendar" development process. First, the designers consider the process of user management, the project improvement plan is in the OSAF design team through the debate and constantly modify the brewing "test drive Method" use, so that programmers better test the code, to ensure that the code at every stage of the correct operation. CHandle's unique design brings a lot of problems, such as how to deal with repetitive events, after three years, CHandle finally began to become some unfinished but still available calendar program, but it is a long time, finally delivered the user can use the software, and name the project name Scooby. Before releasing the bug, get practical feedback, and then try to treat the downtime problem, finally released. Any innovation process is difficult, all need to pay more sweat and painstaking efforts.
Second, personal feelings
In the past, I was not very understanding of some of the methods of software engineering, to its development is also, did not think of its development process is so many twists and turns and difficulties, by reading this last four chapters of the content, let me understand the fact that every use of the method and theory of production are required to practice testing, as the saying goes, Practice is the only standard for testing truth, and every product and standard is hard to come by.
Now, I am more to these predecessors full of admiration, their spirit of the times is indispensable to us, we really should learn this spirit, whether in the present or in the future, this spirit into a power, and strive to explore and improve the software engineering, on the basis of the predecessors of the right to do better and farther.
By reading this last four chapters, I think to solve our current problems, we should learn the spirit of the book programmers to explore, I think in fact, although we are now standing on the basis of previous development projects relatively easy, but we still lack the spirit of courage to explore, This kind of spirit is not what era can be lacked!
Dream Break Code Read Note 03