Why did we succeed in Linux, not BSD?

Why did we succeed in Linux that year, instead of BSD-general Linux technology-Linux technology and application information? For details, refer to the following. It has been quite some time since we started to touch Linux and occasionally try to extend our reach to FreeBSD. I don't know where I came from. I have heard people say that BSD is based on Linux in terms of functionality, stability, security, and so on, even Microsoft Windows 2000 and Mac OS X use BSD program code. In theory, BSD should be the dominant operating system in Free Software. But in fact, it is Linux, which is really awesome, and BSD is more unpopular.

What makes such a difference? Is it true that we don't need BSD, so we 'd rather use Linux to wait for the goods this time? This statement is quite negotiable. BSD must be difficult to compete with Linux due to some factors.

Some people say that BSD's legal lawsuit makes it impossible for BSD to compete with Linux. At least the Wikipedia statement is as follows:

BSDi soon had a legal dispute with AT&T's subsidiary USL... This legal lawsuit delayed the development of BSD's Free Software branch for two years, resulting in great support for Linux Kernel without similar legal issues.

However, BSD is about 1977 market share, that is to say, when this legal case (1992) occurs, BSD has been developed for about 15 years. If we say that BSD was unable to compete with Linux because of the pause of two years, it would be difficult to convince us: is the growth of Linux within two years comparable to the development history of BSD in 15 years? At the end of this legal case (January 1994), Linux even version 1.0 was not released yet! Is Linux (<1.0) more competitive than the old BSD? This is clearly not the problem.

For this reason, I have been Google for a long time and have found the following several arguments. Although they may not be necessary, the combination of them determines the BSD and Linux's survival of the fittest. However, it was too far away from people in that age. Therefore, many of these statements were just as well-spoken. Please treat it as a cool meal!

* Timing:

Linux was around the time when the BSD lawsuit and the Internet took off.

Linux is Linus Torvalds. At that time, there was no free Unix available on the 386 PC, so he decided to write one by himself. Linus Torvalds once said that if there was already javasbsd or HURD, he might not develop Linux. Published in December March 1992, which is about half a year after Linux (December September 1991), and HURD was released only after December 1998. In other words, if javasbsd and HURD can be launched earlier, then Linux may not appear in this world.

In fact, Richard M. stallman is planning to work with BSD people to develop HURD, but some BSD developers are reluctant to work with GNU, the delay in this cooperation case and the final breakthrough gave Linux development and growth space.

At this time, when the Internet emerged, Linux developers and enthusiasts were able to publish news, post new ideas, ask questions and discuss, deliver program code, and return errors in real time through the Internet, this distributed Internet cooperation brings unlimited vitality to Linux. Coupled with the unique management philosophy of Linus Torvalds, many top hackers are willing to participate in the development of Linux. Unlike some large projects, disputes continue and finally fall apart. In this way, Linux has helped the Internet grow at an astonishing rate. When BSD solves the lawsuit two years later, the quality of Linux is enough to compete with those BSD Free Software branches.

* Hardware support:

At the time of Linux's appearance (September 1991), it was just when people started to afford their personal computers. However, the BSD at that time did not support the i386 hardware used by the personal computer at that time, and the general people should not purchase high-price server devices to play with BSD. Therefore, especially for poor college students, if they want to play Unix, Only Linux is available. BSD is not as attractive as Linux.

That is to say, the next generation of Hackers and Lovers are mostly on the Linux side, and BSD is naturally difficult to compete with Linux, while the later version of javasbsd and its successor FreeBSD is the BSD Branch developed for i386 hardware, however, it seems difficult to compete with Linux due to the missed opportunity and the following reasons.

* GNU's strong support:

I think the author of the GNU great program does not need to introduce it more. GNU provides a variety of necessary components required by an operating system, but the most important component-Kernel has been delayed. The development of HURD planned to become the official GNU Kernel has not been smooth, and the emergence of Linux just happened to fill the most important hole in the GNU puzzle.

In addition, although the quality of GNU software is beyond doubt, the people of BSD (especially in the core tool program) do not seem to like GNU software very much, therefore, people who love GNU software seem to have no choice except Linux.

In this way, GNU and Linux work together. Although Linux is just a startup Kernel, it must, in a timely manner, combine GNU, which lacks Kernel to form a high-quality and fully functional operating system. If Linux is so popular, I really want to thank GNU for paving the way for years!

In my opinion, Linux and GNU are inseparable: Without GNU, Linux without any tool programs is useless. Without Linux, the GNU software has no stage to play. Therefore, individuals can accept that the Linux full name should be GNU/Linux. If we think about the development and growth of Linux, I personally think this is not too much.

* Churches and markets, research and hackers:

The most commendable part of BSD is "rigorous 』. Traditionally, BSD has a development Team named "Core Team". The development work, guidelines, and release schedule of BSD are controlled by this Core Team.

However, GNU/Linux is much loose. Linux Kernel is the responsibility of Linus Torvalds, while GNU is the responsibility of Richard M. led by Stallman, each Linux release has its own different support teams. Each tool program and even each function library have different development teams, there are very few communication and coordination mechanisms or the concept of coordinated operation between each other, so they feel loose.

That is to say, BSD adopts a disruptive academic route, while Linux represents the spirit of market-style hackers.

In people's minds, the "rigorous" development method is the road to success, but the appearance of Linux subverts this traditional view. No one can take control of this, so there are any novel ideas, suggestions, criticisms, comparisons, debates, and dissatisfaction in GNU/Linux... Just like the hot pot of boiling water, the water is always surging up: as long as there is something to say, there are always people who listen to and discover the problem, there are always people who are eager to solve the problem, and the comments can always get immediate response. In this environment where the boundaries between developers and users are vague, mutual cooperation, competition, and encouragement, the progress of Linux is of course a thousand miles away. Linux does make the best use of Internet features.

Compared with BSD in the same period, BSD is much lower-key. Although the "Core Team" development model ensures that the quality of BSD is maintained at the highest standard, on the contrary, criticism such as conservative, rigid, rigid, hard to communicate, and hard to attract developers is easily triggered. Although the Core Team of FreeBSD is more lively and open, it is very willing to accept users' opinions, but it still seems difficult to compete with the market-style Linux.

* Various versions:

The loose structure of Linux is also reflected in the release version of Linux. Linux does not have any official releases. Anyone who is interested or competent can independently release Linux. This allows us to find more than 200 Linux releases on the Internet, the actual number is far more than that.

This makes the Linux distribution kit on the market a great deal of competition. Each Linux distribution kit has its own characteristics and appeal. As a result, almost everyone can find a Linux distribution suite that meets their needs. If you are more rigorous, you may choose Debian GNU/Linux. If you want to directly use the full desktop environment, you may choose Ubuntu. If you prefer to compile it, you may choose Gentoo, if you want commercial support, you may choose RedHat... However, some Linux release kits that developers are unable to maintain, have poor quality, or have a small user group may soon be replaced by other Linux release kits. This competing approach also enables the quality of Linux software to be improved quickly.

The Core Team is solely responsible for each BSD release, so at least the quality of the BSD release can meet the requirements of the Core Team. In this way, the system architecture of the people who use the BSD release will be the same in the world, which makes it easier for users: sometimes, even the most basic suite management methods of Linux distributions are quite different, but there is no such problem in BSD.

On the contrary, the result of over-control and lack of competitors is that BSD is not as flexible as Linux. Since users can easily find their own Linux distribution kits, and in addition, Linux may not be worse than BSD, users will not be BSD for BSD, forcing yourself to accept the BSD style, it is not surprising that BSD will become a niche market.

* Richard M. Stallman, Eric Steven Raymond, and Linus Torvalds:

Speaking of the representatives of the free software industry, I think the names listed in people's minds should be the indicators of Richard M. Stallman, Eric Steven Raymond, and Linus Torvalds. Richard M. Stallman is recognized as the spiritual leader in the Free Software field. His opinion is still influential to GNU. Eric Steven Raymond is the Evangelist of hacker culture. He published many articles that have a profound impact on hacker culture, such as churches and markets and wisdom of inquiry. Linus Torvalds is the leader of the Linux Kernel project. These metric figures seem to disagree with each other, but they have a common point-they are Linux advocates.

That is to say, while several representatives of the Free Software Community are trying to promote Linux, BSD naturally disappears from people's radar range. No matter how good the BSD is, what people do not know is also unknown.

* GPL vs. BSDL:

Richard M. Stallman is the spiritual leader in the Free Software field. Apart from initiating the GNU program, I personally think that the GPL he wrote for GNU is a decisive factor. GPL is a feedback term biased towards developers: users can freely use the GPL program code, but all modifications must also be made open to GPL, so that everyone (including the original program designers) can benefit from it. This is the ultimate way to ensure that the program code can always be freely used by everyone.

In contrast, BSDL should be a kind of free authorization for users: programmers do not have the right to declare that the software is authorized by BSDL. Therefore, BSDL software may one day turn into closed software-like Microsoft's adoption of some network components from BSD in the Windows 2000 core, but BSD people have not benefited. Microsoft does not have to give back the modified program code.

This may be the decisive factor: BSD uses BSDL, while Linux uses GPL. Richard M. both Stallman and Linus Torvalds have sternly criticized BSDL, believing that BSDL will tend to split the project, because people can get BSDL program code and refuse feedback, while GPL can make projects more integrated, because we can integrate those GPL program codes into the original program. Therefore, many top programmers insist on developing software only with GPL, so they do not want to spend time to marry. It is unexpected that BSD using BSDL is left empty.

Therefore, the loss of BSD in the past is not surprising because many subjective and objective factors are conducive to Linux. As late as today, BSD seems to have no momentum to make a comeback. For example, in the top 500 super computers, Linux accounts for 427, Windows 5, Mac OS 2, and BSD only 1. Although it seems unfair to use this method, we can see a little clue from it.

Here, you may be curious: FreeBSD is a BSD Branch developed for personal computers. FreeBSD has been developing fast in recent years, taking a picture of the excellent tradition and outstanding reputation of BSD, is it possible that FreeBSD will squeeze Linux into the new overlord of Free OS one day?

In fact, FreeBSD and Linux only have Kernel and Core Tools, and the difference between other application software is small. That is to say, FreeBSD's real competitor should be GNU + Linux. So, based on the arguments in this article, this will be a FreeBSD vs. GNU/Linux, church vs. Market, study vs. Hacker, BSDL vs. GPL war. However, the history tells us that such a BSD path has failed. Before some objective factors change, does FreeBSD want to reverse Linux? I personally think it should be not that easy!