Wei Yongming: They all want to build together in the name of open source collaboration, but there are many hills and mountains.

Wei Yongming, Bachelor and Master of Engineering, Tsinghua University, founder of Feynman Software, outstanding contributor to open source software. The well-known open source software MiniGUI was released in 1999 and has been continuously developed to this day. He has published technical works such as "Analysis of Minicul" and "Linux Device Drivers" (second and third editions). In November 2018, the Open Source Collaboration Project of Hebi Operating System was initiated. In August 2020, Lead Street wrote and published the first domestic novel "Kao Ding Ji" in a farming style and published it online. In August 2020, proposed and developed a new programming language HVML.

This article was first published from Wei Yongming's public account: Test , the following is the full text of the article

In the "161c40444ac8ed talk less about feelings and show less, write more code and innovate more ", I criticized the phenomenon of deified open source. Today's article criticizes the phenomenon of deification of open source collaboration.

Some people say that "open source collaboration" is an innovation. I don't agree with this view.

Open source collaboration can only be regarded as an innovation in the cooperation model at best, but I don't like to use the term "at best". Because emphasizing model innovation will bring many negative effects. Just like the hot spots of the Internet in the past few years, such as shared bicycles, they are all products of model innovation, but whether these model innovations are good or bad for the development of society is still inconclusive and remains to be seen.

In addition, many people’s first reaction after hearing the phrase “open source collaboration is an innovation” is “open source collaboration can promote technological innovation”. But what I want to emphasize is that open source collaboration does not promote technological innovation.

My point of view, in plain and easy-to-understand words: open source collaboration is suitable for conservation, not for development.

Let me explain to everyone next.

First of all, I am not denying open source collaboration and organizations that promote open source collaboration, such as the existence of various open source foundations and open source communities, which have their own value.

open source collaboration is a collaborative development mechanism based on a certain open source software. It is essentially the same as the technology alliances and standards associations that have existed for many years. It is just a mode for people in a specific circle to discuss some things. To effectively carry out open source collaboration, currently two platforms are mainly used. One is a virtual communication platform built on various Internet tools, such as mailing lists, technical forums, etc., which can be called open source communities. The other is the Open Source Software Foundation.

Similar to technology alliances and standard organizations, open source collaboration platforms, whether in the form of virtual communities or open source foundations, essentially have value when a technology enters the promotion and maturity period. is that some people who represent different interests use this platform to balance their interest relationship , such as two similar technical points, everyone fights, which should become the standard. Change to open source software, that is which module or which protocol can be put into the open source software.

If we analyze the open source software that currently plays a pivotal role in the entire computer industry, you will find a rule: These open source software are often open source copies of commercial software. These open source software are either developed by individuals or organizations and then managed by the foundation, or open source alternatives to some existing commercial software developed by open source collaborative organizations. like the Linux kernel, is a substitute for the commercial Unix kernel. Various tools and software of GNU are also a substitute for commercial Unix systems. LibreOffice is a substitute for Microsoft Office. FireFox is a substitute for Internet Explorer. Only a very small number of open source software, especially software developed by a commercial company, can be considered as more original software, such as Android, and some free software based on dual licenses, well-known ones such as MiniGUI, MySQL, MongoDB, etc. (If you look closely, Android, MySQL, etc. are actually open source alternatives to commercial software.) This part of open source software rarely mentions open source collaboration. For example, Google has absolute control over Android, while free software based on the dual-license model needs to ensure that the copyright ownership is clear and exclusive. It can be said that the vast majority of open source software, especially the products of open source collaboration, have little to do with technological innovation or progress. And as long as it is original software, whether it is open source or not, it must be related to one or two key figures or teams (such as Linus for Linux kernel, Mozilla Foundation for Rust language, Microsoft for TypeScript language, Wei Yongming for HVML language, etc.), It can be said that open source collaboration is not the original driving force of these original work.

Of course, widely used open source software such as the Linux kernel has many innovations in the course of decades of development, but in general it can only be regarded as a partial repair and cannot produce fundamental changes. In terms of operating systems, FushiaOS and HybridOS have brought fundamental changes. But in the beginning, open source collaboration often missed them-for an innovative technology, at the beginning, most people's attitude was just wait and see.

In short, others first have innovation and then have an open source alternative, or first have innovative things and open source, and then they can talk about open source collaboration. The main contribution of open source collaboration to technological innovation is the application that can accelerate technological innovation. But this is another topic, we will talk about it later.

The reason is also very simple: innovation is a costly economic activity that requires long-term accumulation and investment; and only benefits can become the first driving force of innovation. That kind of open source philosophy based purely on dedication is not enough to support people's long-term and dedicated investment. Therefore, lazy people only need to copy the ideas and creativity of others, write a few lines of code to open source, and then hope to magnify their reputation through the Internet, and then get the rewards they want. Even as for the existing open source software, you can change the name to become your own.

Specific to our country, this problem is even more serious.

Chinese emphasize pragmatism. Few companies have a long-term vision and can make long-term investments in innovation. If everyone does not see the future of a technology, it is impossible to build a platform to engage in collaboration. This is human nature. Therefore, collaboration can only be established when the original technology is partially recognized, and the collaboration platform can be used for development during the promotion and maintenance periods. In addition, the main participants in this type of collaboration platform are companies, and the primary purpose of companies is to make money. If you don't have money to invest, most Chinese companies won't do it. Even if you participate, in most cases it is just a gesture. You see, our country has had many technical alliances for so many years, and in the end they either fell apart or ate and drank without much substantial output.

In addition, currently has only one open source foundation in our country, which is the Open Atom Open Source Foundation . Huawei, Alibaba, Tencent, etc. are all sponsors of this foundation, and everyone donates various open source projects to this foundation. For example, operating systems include OpenHarmony, openEular, AnolisOS, and so on. Among them, openEular and AnolisOS were originally a competitive relationship, and now they are donated to the Open Atom Foundation. Think about it, whether the foundation has the ability to develop so many operating systems at the same time is one thing. In terms of these two competing operating systems, how does the foundation balance the interests of the funders behind it? We see that these operating systems donated to the foundation are, in the final analysis, a system built on the existing open source software technology. At the beginning, there was not much technical innovation. Putting it on the same internal volume platform, can innovation be produced?

As such as a foundation, it is very easy to become an platform or a bureaucracy 161c40444aca63. Take the world's most popular Linux Foundation, a lot of funds have been raised in the past two years, but there have also been many weird projects. For example, the SPDX specification, whose full name is "Package Data Exchange Specification", I learned about it, and it probably regulates everyone's open source license declaration and use. Projects like this, in my opinion, are supported by a full meal—just finding something to do for a large group of bureaucrats raised by the foundation.

You see, on top of open source software that does not have many innovative genes, you can expect to innovate through an involuntary, mediocre, and bureaucratic open source collaboration platform? Isn't it a bit idiotic to talk about dreams?

Therefore, my view is that open source collaboration does not promote technological innovation. Technological innovation comes from arduous efforts, deep insights into products and technologies, and long-term continuous investment. It is wrong to rely on open source collaboration for technological innovation.

You, it's time to wake up!

PS: The latest news before the publication of this article: Another OS similar to openEular and AnolisOS called OpenCloudOS has been released. Hey-hey……

---

content of this article is authorized to reprint by the author, and does not represent the platform's views.