1. Foreign open source communities

The first and most mainstream group of open source abroad is community. Linux kernel, Debian distribution and GNU open source library are all very pure community models.

Second is the commercial organization, through open source, enhance the influence and status of the enterprise in the community. For example: OpenJDK for Java, Google’s Android

There are also startups that have become popular through open source, such as Docker and Spark.

Foreign universities are also a major contributor to open source, but they are co-managed by community organizations. Berkeley, for example, is a grandmaster in the open source world.

But what they all have in common is the need to rely on the community as a source of nourishment.

In our own country, the community is not a climate, it is a corner. On the contrary, by virtue of their deep pockets, Internet giants continue to contribute excellent open source products to enhance their influence in the industry. So this open source route, I understand is the commercial provisions of good open source, it is difficult to have a sustainable prosperity of the basic guarantee!

2. Analysis of slow domestic development

So why is our open source community growing so slowly?

On the one hand, domestic technical engineers are late starters, making it easier to access useful existing open source components rather than building wheels. Domestic engineers are more accustomed to blog in the form of words to summarize technical experience and form personal influence, because this kind of investment cost is much lower than the establishment of open source library.

On the other hand is our social culture, for free things like to ask, do not know how to feed back! Engineers treat open source more as an integration of their own projects and less as a technical contribution back to open source. In addition, the team consciousness without interest drive is very poor, and it is difficult to form interest-oriented team collaboration.

On the other hand, although the salary of engineers is relatively high, but the social status is low, most business owners do not know how to respect technology and talent from the heart, but squeeze too much, 996, Wolf animals! Engineers have less time to create open source libraries.

Finally, we don’t have any major open source universities, not even close. And domestic capital is quick money, it is difficult to make long-term investment in open source.

3. Why is it difficult to create an open source ecosystem like Hadoop and Spark

When you look at the Hadoop source code, you realize that it’s so engineered that it can’t be done by one person working part-time at home. It takes a team dedicated to it.

Hadoop was the brainchild of Doug Cutting who worked at Yahoo, Hadoop is currently supported by the Apache Foundation, and it’s being commercialized by Cloudera, and Doug Cutting is currently at Cloudera, Cloudera, in turn, was co-founded by former employees of Google, Yahoo, Facebook and Oracle.

Spark was originally developed at Berkeley Lab before its founders, Databricks, secured financing from an investment firm.

This kind of support for open source can be seen by technology giants, foundations, universities, American companies, venture capitalists and technology startups. This kind of support is three-dimensional and multi-dimensional, and Hadoop and Spark are basic software, and it takes a long time, effort and investment to make money by themselves. We may have different goals, and I’m sure it’s never just about money.

Look at our country’s science and technology enterprises, how many open source projects have been taken by colleges and universities? Is there an ecosystem of open source projects? Are there any well-supported open source foundations? Are they tired of 996’s blessing, or can enterprises spare 30% of their time for Danniu?

I still need to borrow a comment from People’s Daily:

Internet giants, do not only think about the flow of a few bundles of cabbage, a few kilograms of fruit, scientific and technological innovation of the stars of the sea, the infinite possibilities of the future, in fact, more exciting.

Well, for tech bulls to ignore external support and put together a team to work on an open source project on their own, I believe there is a 99.9% chance that investors will not be impressed before they starve! The time cycle required for basic open source software is simply unaffordable for individuals!

Do a good job in the enterprise part-time, can, such as successful, Musk is a Martian!

Although big companies like Ali have made great contributions to open source by creating Canal, MongoShark, Dubbo, Druid, Fastjson, Fastdfs, TFS, Rocketmq, OceanBase and other excellent open source projects. (For the record, I’m not discrediting alibaba’s recent reputation.)

But the problem is that Hadoop and Spark are the difference between the ecosystem and the tools that Ali has contributed to open source, and I don’t know if you understand that, but they’ve created a platform, an ecosystem, a technology world that can suddenly make RocketMQ disappear, no problem, Kafka Rabbitmq comes on top, But can Hadoop and Spark suddenly disappear? This leads to systemic risk!

4. To summarize

But on the whole it’s better. Remember that ten years ago, the place with the least contribution to open source in the world was China. However, most of the popular open source databases in China today still come from commercial institutions. For example, the open source databases made by start-up companies like TiDB can be described as a complete exception. Therefore, I think it is an effective way for the domestic open source to continue to prosper by relying on the continuous efforts of technical entrepreneurs in the field of open source.

The open source community culture is really far away and will probably never thrive. Unless our society has a fundamental understanding of open source, and of technical talent to engage in open source, don’t play with the empty, with the real thing to support. Through entrepreneurship drive open source community strategy, form interdependence. Such ability turns the development momentum of domestic open source thoroughly.

Open source projects are especially easy to form the basic software ecology, which is an important weight in the game between countries. However, the American empire will not die, and the basic software will be magnified infinitely as our weak link.

Especially open source software, in addition to the strong support from the state, only in various fields, not only in the field of science and technology, those experienced leaders make joint efforts and continue to create, it is time to stimulate the creativity of the leaders, don’t wait any longer!

It is hoped that far-sighted universities can provide more support and help to technology entrepreneurs, so that they can work together to shoulder the banner of sustainable development of open source. For the experimental subjects of colleges and universities, if supported by the cooperative open source projects of entrepreneurial teams, it will certainly form a hard core force to promote the development trend of science and technology. Isn’t this the integration of production, learning and research that we have always wanted? It could even install a durable economic driver for society.

The article is from the public number read byte (focus on distributed architecture and big data technology) original. If reprint this article, please be sure to indicate the source of the public number, thank you for reading and forwarding.

Read the bytes | creative center

Head over to read Byte Creation Center – learn more about Read Byte creation