Talk about technology, talk about operation, talk about ideal

I haven't updated for a month, because I've been too busy recently, and I can't remember how many overtime I worked all night this month! However, the two biggest problems in the system and operation have finally been solved, and it is not a waste of time!
First of all, let’s talk about some self-adjustment problems! Since last year, until this year, I have been immersed in my work day and night, and I admire myself for persisting until now. I am not a superman, but also a mortal. Saying that I don't have some mental fluctuations is a nonsense. For example, when I couldn't see the path ahead at first, I thought about whether to give up, but in the end I insisted on writing there like a fool; The reason is that I almost quit, because I usually only see me when I get up, and I can't give them anything back. During this period, I would also read the Analects of Confucius to learn from some things and adjust myself. When I encountered work pressure, I would also release it with my friends. Anyway, working hard, being just and magnanimous, and getting along well is definitely not wrong.
Let's talk about operational issues! Speaking of the current operation project, there are thousands of words that are unclear, and I will not go into details. I mainly talk about some of the experience in operation summarized in this project. First of all, this project is a cooperative project, and we also invested in it, but last year, due to the lack of expectations for the difficulty of this matter, and our own focus on our own products, we did not have the energy to take care of it. This matter is arranged by the people below. Just do it, and the problem is obvious. First of all, there is no clear decision-making and implementation plan, and the responsible person and team members have basically no contact with this industry. They are all newbies. How can they do well without paying tuition! For our existing products, we paid more than half a year of tuition and took countless detours to get on the right track! Secondly, the people at the top of us may not be wrong in controlling the overall direction, but it is very difficult to do a good job without one person standing up to do the relevant program research and deployment in person. Many things are not done in person. Touch, you won't know where the problem is at all. Moreover, training and inspection must be done in the early stage. If the plan has not been implemented in place, it is equal to zero.
This year, I have almost finished working on the product at hand, and I started planning this matter myself at the end of March. Since I have a clear understanding of this matter, plus some advantages in technology development, I quickly came up with a decision. And the implementation plan, but I still did more than 2 weeks of testing research and adjustment myself, and repeated polishing during the testing and research process, often in the middle of the night, after doing it, I did staff training, and did batch running tests. Recently, all aspects have been relatively smooth, and I believe that the data will reach the established goals soon. To sum up, to do a project, you must do pre-research and make plans in person. If you don't have the energy, time and relevant experienced people, don't do it directly, otherwise the ratio of failure to success must be 8:2. If it is a technical class When doing this, operations must not make mistakes in technical operations, such as information blockage, unwillingness to learn from communication, unwilling to self-correct, obsessive-compulsive pursuit of their own unnecessary extremes; but also not like non-technical operations. Like personnel, the goal is to be achieved entirely by human resources tactics. Everything must be based on data indicators related to project operating income, such as maximum revenue flow and automation within an acceptable time frame, and the lowest labor operation cost.
Let's talk about technical issues, after all, he is still a cute programmer! First of all, the problem of system distribution was finally solved in the past month. Before, because a functional module in the system was not designed for distributed design, it has been unable to do more distributed deployment. The colleagues in charge of this module have more work on their hands. In order to speed up the progress of the product, I wrote it myself. I usually have to deal with a lot of things, and I have time to write every night. So I wrote it for about 2 weeks, and I finally got it done. I thought it should be no problem. As a result, there were still 2 serious bugs during deployment, and they were not detected at all during small batch testing, but in fact, it was not caused by deployment, but the problem of the client. One is caused by the improvement of the authorization function, which makes the authorization of a large number of users invalid. The conclusion is that the stable code must not be touched again. Every improvement must be accompanied by a toss!
The other is that some scripts do not throw exceptions and cause some infinite loops. When it comes to exploring this problem, it is also quite painful. Since the APP was written entirely by myself at first, it took me about 2-3 months to write a stand-alone APP. Due to the more and more complex functions, the code was written in a mess, and now other colleagues may not be able to I understand 100% of the code logic in it. As a result, it is difficult to locate a problem this time, and there is no error log, nor can it be restored immediately! As a result, I spent the night recalling the code, and slowly searched and found it, and finally found it. To sum up, the code must be concise and clear with more comments!
Finally, let’s briefly talk about the summary of management! Recently, I have a total of 3 R&D and operation projects in my hands. In fact, the work is about focus. Although many projects can be solved by adding manpower, experienced decision-making managers are very scarce, so they can play freely. It is often difficult to achieve success, at least none of the projects that I have experienced have been successful. Therefore, it is very important to cultivate a team. After all, a person's energy is limited. For example, an existing project has been relatively mature, and he has gradually cultivated and separated it for project members to do it, and only do some demand follow-up, and there will be similar follow-up. project, you won't have to worry about it anymore. However, some projects that are still in the semi-finished and beginning stage are very laborious and laborious, and some may not have much experience themselves. This kind of project is even more challenging, but if you have successfully completed a project, do other new ones. The project practices are also similar, such as doing more program testing and research, collecting more relevant data and research, communicating with people with relevant experience, and learning from some good points of similar products. Finally, as the team grows larger and larger, we must make full use of the existing team resources so that every member can gain improvement and relevant experience, otherwise, there will be more and more things and projects in the future, and it will gradually become It is found that there are always only a few people available. Although there are many people, the quality has not improved. The more capable and experienced people, the higher the probability of a project's success.