This iteration seems to have completed most of the work very early. But in the last day, it is still quite busy.
Modified a lot of fine errors. Lessons: To set the quality standards in the beginning, and check it regularly. Otherwise, it will never reach this standard.
The user manual is finally not completed, and the example is missing. And it is not very satisfied with the quality. Lessons: Even if the document should also set up a measured standard, and divide a small task for no more than 3 days.
The above two problems are due to this iterative amount of my work. A wide range of reconstruction makes my main energy into the top, and it is not completely completed yet. Lessons: For technical leaders, the most important task is: understanding problems, managing exchanges, management quality. In addition, significant reconstruction is not wise before important delivery. Further, the complete test allows you to stop when you want to tell a paragraph, and will not let the change dragged you into the quagmire of the code. Further, the change should first implement the function, even if the hardcoding or duplication code is not hesled, the structure is gradually corrected in the subsequent reconstruction. Unfortunately, another route I use, making the difficulty of work increase.
It feels that this reconstruction is required in the previous development. At present, as long as you write code, there will be structural defects. Haven't found a good way to ensure that the overall design quality remains in a stable level. I feel that the problem is not enough speed, the opposite is too fast.
However, the release is still fundamentally satisfied, and at least the team has provided an opportunity for the preliminary work in the team. It is also possible to formally obtain a customer's needs.
The next version of the function definition is required later. Consider using Cooper's example users, and formally listed the work written by the user files in the task list.
