In the past two days, there has been a big discussion in the technology circle about the localization of databases. Are domestic databases really self-developed or are they used as shells to make money?
Feng Ruohang, founder of the open source PG distribution Pigsty, published "Is the database really stuck?" on the WeChat public account "Illegal Jia Feng". " The article believes that many "domestic databases" are defective products that cannot be supported by mud. Package open source free software and sell it to you at Oracle prices!
"Really self-developed databases are polarized: a very small number of products that truly have innovative contributions and practical value cherish feathers and do not deliberately advertise as "domestic." And most of the remaining ones are often indigenous databases that are built behind closed doors and have backward technology. Or the low-quality wheels of open source that were forked and negatively castrated in ancient times. Domestic databases are not good companies without solid work, but the label of "domestic" has been contaminated by a large number of mediocre and inferior products that have penetrated into the database field.
"What's even more frustrating is that bad money drives out good money. Such squandering of already scarce database R&D manpower will actually choke the domestic database industry. Especially in the core OLTP/relational database field due to the existence of open source , there is no shortage of kernels that are useful enough. Being able to make good use of PostgreSQL/MySQL and provide service support is far more valuable than the self-deceiving Dalian kernel."
Feng Ruohang believes that there is no shortage of perfect database kernels in the industry - such as PostgreSQL, which is fully functional and open source and free (BSD-Like). Countless "domestic databases" are based on PG's magical modifications. If anyone got stuck on the database core, it was definitely because he was full and choked.
But database localization is necessary. As the article "Is database localization a trap to make money" published by the WeChat public account "White Eel's Cave" said: " Once you are on the blacklist of the United States, you want to buy Oracle. It’s not for sale either.”
The article also believes that using domestic databases to make money is really a bit too imaginative.
"Those who want to make money don't have to work on databases. There are a lot of projects that can make money better. Most of the friends I have met with domestic databases still have some sentiments. Just like ten years ago When I met Cui Weili, the then chairman of Nanda General Motors, I had the same feeling when he said he wanted to do something related to Tianyuan. Only people with a little sentiment would do this kind of long-term business where they don’t know when they can make money. Buy and sell.
"There are many people I know who are very successful in the database field. Since I joined the circle to self-research databases, although I have seen them raise hundreds of millions and hundreds of millions, but after all, these people are basically doing the same thing. It’s a job for a passing God of Wealth, and the money raised quickly goes into a bottomless pit of R&D and marketing. Despite this, the salary of technical personnel in domestic database companies is not high, except for a few companies with deep pockets and long-term investment. , the R&D backbones and senior management of most domestic database companies do not make as much money as some people imagine. The theory of making money is actually untrue. What I see is that there are many domestic database practitioners. People who have been in poverty for ten years are still obsessed with it. Even if we don't respect such people, there is no need to slander them."
We have been talking about going to IOE for many years, and we really need to put it into practice, and it is not just a matter of shouting slogans. Xue Xiaogang published an article on his public account "Brothers in the Four Seas" titled "The Cost of Replacing Databases and Authentic and Fake Domestic Databases" that from the perspective of development, operation and maintenance, and cost, it is not easy to abandon Oracle and replace it with domestic databases.
"There are also good ones made in China, maybe only a minority. These are concentrated in those who were engaged in this industry before 2016, and they may be really doing it. It is okay to do it when it is not favored. At least it is not a lie. As for whether it is done well or not, that is another matter. In short, it is not just a matter of grabbing money that comes out after the policy is established. But if everyone does it in a rush, it is basically chaos.
" Everyday I see many people using Apple phones, MAC laptops, driving Teslas or using IPads, and they tell me that they want to replace a certain domestic database. These people may not be involved in development, operation and maintenance, or maintenance. Participating in procurement means standing up and talking without pain. Who pays for these costs? Will you be responsible if there is a problem? Many people in the chain just talk about database selection without knowing anything. Isn’t this a layman commanding an expert? "
Regarding the localization of database replacement, do you think it is necessary? As for the way to localize it, should we re-develop the database core, or should we localize it based on the existing open source database? In fact, it is not difficult to tell whether the database is self-developed or not. Why are there so many companies trying to make money? Leave your opinions in the comment area~
Alibaba Cloud suffered a serious failure and all products were affected (restored). Tumblr cooled down the Russian operating system Aurora OS 5.0. New UI unveiled Delphi 12 & C++ Builder 12, RAD Studio 12. Many Internet companies urgently recruit Hongmeng programmers. UNIX time is about to enter the 1.7 billion era (already entered). Meituan recruits troops and plans to develop the Hongmeng system App. Amazon develops a Linux-based operating system to get rid of Android's dependence on .NET 8 on Linux. The independent size is reduced by 50%. FFmpeg 6.1 "Heaviside" is released