Today is October 4, Mid-Autumn Festival, according to the date on the ID card, also happens to be my 49th birthday. My wife took my son to changchun to participate in the national fencing club league, and I was alone at home. When I woke up in the morning, I habitually opened my notebook, looked through the code I had written, and made a small optimization, which made the system configuration parameters less. After counting my lines of code, I found that I had written more than 38,000 lines of code in six months, which was more than the total amount of code in ten years of working in the United States.
I have been working as a programmer for 33 years since I started to write Basic programs in 1984 when I was in senior high school. Although I often write programs and like to write programs, I never thought that writing programs was too easy to be specialized in, so I majored in fluid mechanics and astrophysics in university and graduate school. However, after leaving school to study the large-scale structure of the universe, I could not support myself by astrophysics, and eventually entered Motorola and began my career as a professional programmer. In 2007, WHEN I was still working at Motorola, I felt that the spring of mobile Internet was coming, and mobile Internet must have push service, so I should build a push platform to provide professional mobile push service. When he returned to Beijing in early 2008, he wrote about 3000 lines of code and realized a mobile Internet push engine which is still proud of being efficient, energy-saving and flow-saving. At the beginning of 2013, when I saw the rise of intelligent hardware, I started my business again and founded Happy Mommy. The technical threshold was not high, so I didn’t want to write any one-line program, but the R&D team couldn’t figure out how to calculate the fetal heart rate. At last, they used the method of analyzing the large-scale structure of the universe and calculated the fetal heart rate accurately in real time with 600 lines of code.
In my opinion, programmers write programs, like painters paint, to create work, if the work can be appreciated, that is the greatest reward. Since 1984, I have developed countless software, but the most let me proud of is two, one is 1993 summer vacation with FoxPro to help sister develop financial software, 20 years later, to today, sister is still using, open the software, the above display is still “Jianhui computer”. There is another software, THE LB membrane balance system developed by ME in the polymer LB Membrane Laboratory of USTC when I was studying in The University of Science and Technology of China. Its hardware and software are still running every day, and the data of hundreds of SCI papers are from the LB membrane balance developed by myself. Now when I open the system, “Tao Jianhui” is clearly displayed on it. These two software, did not let me get financial returns, but let me most satisfied, there is unspeakable pride, because it is still in use, use very well.
From school to work, he was successful, pretentious and considered himself a technical expert. But if you think about it, this tech whiz has a lot to lose, because I’m not a contributor to any of the world’s most popular software. Fortunately, I haven’t reached the point where I can’t write a program. I still have a chance. A year ago, I realized that there was a niche market for sequential databases. Database is the underlying basic software. Different from mobile applications and websites, it is slow to change and has a long life cycle, but the technical threshold is quite high. On the other hand, the traditional database has been studied for decades, and all kinds of indexes and storage structures have been thoroughly studied, so it is difficult to make a difference in technology and market. However, due to the development of mobile Internet, the rapid growth of data volume, single machine has been unable to process, must rely on the cluster. The core technology of database is now how to design distributed system, processing high concurrency, efficient real-time data synchronization, query, calculation. Since my first day at work, I have been working on distributed systems in the communications industry. The systems I designed have always supported hot swap, online upgrade and no single point of failure. These experiences and techniques can be applied to the design of distributed databases. So I decided to become a programmer again and started writing programs.
After a few days of setting up the development environment and familiarizing yourself with the development tools, we were ready to go. People close to 50 would have thought it would be a lot less efficient than it was 10 years ago. But unexpectedly, with a clear goal, my potential was fully tapped. Since the middle of December 2016, I worked more than 12 hours a day on average and wrote more than 18,000 lines of code for two months, realizing the core engine of the entire timing database. A simple comparison test found that the performance was at least 10 times faster than MySQL, MongoDB, Cassandra, Influx DB, Open TSDB, etc. Excited by such results, I decided to officially register the company, raise funds, set up a team, and once again walked on the road of entrepreneurship.
After getting investment from Xue Manzi and Mingshi Capital, TAOS Data officially opened in June, and immediately there was a r&d team of several people. I was no longer fighting alone. But in order to ensure quality and schedule, I didn’t just want to do directing and so-called architectural design, but to work with everyone and focus on writing code. Three months later, I found myself contributing another 20,000 lines of code, turning a sequential spatial data engine into a testable product, TBase, a big step closer to a commercial product.
In the past ten months, I have been writing programs and fixing bugs almost every day. Whenever a difficult BUG is solved, I am always unspeakably happy and immediately invite the team to dinner. Every time I find another clever way to improve performance a bit, I can’t help but share it with my team. Every time you’re stumped by a problem, you don’t sleep well, you wake up early and want to solve it. More than ten hours of development every day did not make me tired, but let me go back to 20 years ago, the brain can operate at high speed. If I don’t write a few lines one day, I always feel like I have something left to do. Two weeks ago, I attended a forum in Shanghai. I realized the Naggle algorithm easily and passed the test. Immediately, I felt a strong sense of satisfaction and pride.
I am very glad that I made a decision a year ago. Before I get old, I still have the opportunity to develop a top product with considerable technical threshold, and I can use all the technology and experience I have accumulated for many years. And fortunately, xue Manzi and Mingshi Capital angel investment again, set up a capable but very aggressive technical team. Based on TBase’s test results so far, it’s a shoo-in. No matter what the future market promotion, whether open source or commercial, I believe there will be many people like this product and taste the beauty of my design and programming. If TBase is still in use 20 years from now, like the financial software I developed for my sister and the LB film balance I developed for my lab, I will be the happiest old man in the world. If my son proudly told everyone that the core engine of TBase was developed by my father when he was 49, that would be my greatest legacy to him. If my grandson still knows TBase and TAOS Data, that is the greatest inheritance I have given him.
The previous two entrepreneurial experience, let me see enough of the ebb and flow of the clouds. I am best at writing programs, the education, training can let me in this area far better than others, why to engage in pushing, operating, do some surface thriving, and do not need any technical threshold to live? Or continue to write programs, and write well, to write, like a painter, attentively to create a work, and not draw entertainment work. I hope that 10 years later, no matter TAOS Data has been IPO or a small company, I will still actively contribute code to TBase, and 20 years later, I can still participate in the discussion of technical details and solve bugs by hand. The last moment of my life, hope is in front of the computer screen.
No matter how much money there is, it is difficult for people to leave traces in history, but a good work can be passed down and enjoyed by future generations. Let me lead the development of TBase to become a masterpiece, Leave a dent in the world!