America's No. 1 financial conglomerate

Chapter 417 The First Shot in the Siege of Microsoft

Taking advantage of Microsoft's current internal turmoil and focus on lawsuits, Ernst is about to fire the first shot in the operating system war.

In the living room, the television volume was already turned down to the lowest setting. Ernst turned his gaze to Karl Weller. "How's things going on your end?"

Upon hearing this, Carl Weller shook his head slightly, a helpless expression on his face, clearly indicating that the result was not ideal. "We have had two in-depth conversations with Linus. He has shown great interest in joining the Bell Labs research team, but when it came to commercializing his research results, he refused without hesitation, his attitude was very firm, and there was no room for negotiation."

Carl's eyes unconsciously revealed a hint of admiration. "Linus is the purest kind of scholar. In his mind, the significance of technological research and development lies in promoting industry progress and benefiting mankind, rather than measuring his achievements with commercial metrics. He cares more about the value that technology itself can bring to users, rather than how much profit he can make from technology."

"Foolish." Ernst cursed without hesitation after hearing Karl's report, his tone filled with obvious dissatisfaction and disdain.

Carl simply shrugged, not wanting to get involved in commenting on who was right or wrong.

Linus adhered to the pure ideal of prioritizing technology and regarded commercial interests as impurities that tainted the essence of technology. From the perspective of a research scholar, Linus's behavior is undoubtedly admirable.

This kind of dedication to seeking fame without seeking profit is a model for scholars and a state that all researchers aspire to.

In reality, Linus's insistence, just as Ernst said, was utterly foolish.

Especially in this money-driven era where capital is the core driving force, the research, development, iteration, and promotion of any technology cannot be separated from the support of huge sums of money.

A technology that cannot generate profit, no matter how advanced or perfect, will struggle to attract capital investment.

Without capital injection, the R&D team cannot obtain sufficient funds to purchase advanced equipment, recruit top talents, or have the resources to carry out more forward-looking technology projects.

If this continues, technological research and development will stagnate and eventually be eliminated by the market.

Ernst didn't dwell too much on Linus's rejection. "What do you think of the Linu system he developed?"

Linu, which will later become the most widely used, most popular, and most influential operating system in the world, is the core of Ernst's future efforts to shake the foundation of Microsoft.

However, it is now virtually impossible to challenge Windows' dominance in the field by relying on Linux as an operating system for personal computers.

First, there are the alliance barriers at the hardware level. The solid Wintel alliance between Intel and Microsoft, and the absolute dominance of the 86 architecture processor in the personal computer hardware market, create a deep binding between hardware and software, which naturally hinders other operating systems from supporting the hardware.

The second constraint is the software ecosystem. Currently, most software products developed by software vendors are based on English.

The technical architecture of Intel processors and Windows operating systems was designed.

This means that even if a new operating system surpasses Windows in technical performance, without sufficient software support, the computer purchased by the user will face the embarrassing situation of having no software available, and will only be an empty shell.

To completely bypass Microsoft and Intel, we must wait for the advent of the smartphone era.

At that time, the hardware architecture will be completely freed from the constraints of the Intel 86 architecture, providing a completely new hardware platform for new operating systems.

The software ecosystem was first built on mobile devices, and then, with the convergence of mobile internet and personal computers, it directly expanded into the personal computer field.

Only this ecosystem building model, which prioritizes mobile devices and then penetrates into PCs, can truly circumvent Microsoft's absolute advantage in the personal computer ecosystem and catch Microsoft off guard.

"I don't have any particular opinion." In response to Ernst's question, Carl Weller gave an objective assessment.

"From a technical perspective, the Linu system doesn't offer many groundbreaking innovations. It's essentially based on the Uni system developed by Bell Labs and is a type of Uni-like operating system."

"The reason why the Linux system has risen so quickly and become so popular is mainly due to its two characteristics."

"First, its open-source nature means anyone can obtain the system's source code and modify or optimize it; second, its unified kernel ensures consistency in core functionality across different versions of Linux, resulting in stronger compatibility compared to traditional Unicode systems." Regarding Linux, a system revered as a god-like entity in the open-source community, Carl Weller didn't express much...

Highly recommended.

In his view, if we disregard the influence of the open-source community and commercial promotion factors, and judge solely from a purely technical perspective, the performance and architectural design of the Linux system do not fundamentally surpass some of the systems developed by Bell Labs in its early stages.

A pretentious way of putting it.

That's all there is to it.

Ernst frowned slightly and asked succinctly, "What I want to know is, do we need to purchase a commercial license for the Uni system?"

Although it is true that the UNIX system was developed independently by Bell Labs, in 1988, AT&T sold the trademark rights and source code ownership of the UNIX system to Norwell Corporation.

Bell Labs has the right to use it internally, but if it wants to commercialize products developed based on the Uni system, it must negotiate a commercial licensing agreement with Novell.

"I think it's completely unnecessary," Carl Weller said without hesitation, directly rejecting the idea.

He leaned forward slightly, looked at Ernst, and continued, "From a technical architecture perspective, the Uni system has two fatal flaws. First, the kernels differ greatly; different vendors develop derivative versions based on Uni with huge differences in the design and implementation of the core kernel. Second, the system has a high degree of integration; the system's hardware drivers, applications, and kernel are deeply bound together, making it difficult to modify and extend flexibly. These characteristics determine that the Uni system is not a good choice."

After Carl finished speaking, a brief silence fell over the living room.

Because none of the remaining three people could understand what was being said.

Carl also noticed this awkward situation and explained it in simpler terms: "Simply put, the Unicode system is too independent, resulting in extremely poor compatibility. Various Unicode derivatives on the market, such as Sun's Solaris, IBM's AIX, and HP's HP-UX, are all developed based on Unicode, but they are almost incompatible with each other."

"An application developed for the Solaris system cannot run on the AIX system; the file formats and command interfaces of different versions of the Unix system are also very different, making it impossible for users to operate across systems. But Linux does not have this problem. As long as the kernel is Linux, no matter how it is developed, as long as your application can be adapted to one Linux derivative system, it can be used on almost all other Linux derivative systems. Even if it is not a Linux system, it can be used with simple modifications at most."

"Ernst proposed developing a source code management tool, not for his own use, but for commercialization to unify the entire market. This tool must be highly flexible, adaptable to different operating systems, support different programming languages, and meet the personalized needs of different enterprises."

"Compared to the Uni system, using the Linux system as the kernel base is clearly more in line with our needs."

"We can fully leverage the characteristics of a unified kernel and strong open-source attributes to carry out customized development, while also relying on the power of the open-source community to attract more developers to build an ecosystem for it."

The three of them seemed to understand this time, but Hamilton immediately raised a new question: what exactly is a source code management tool?

He is not a technical expert like Sergei Brin or Larry Page; he studied business in college and has a very superficial understanding of these kinds of technical issues.

Carl looked at his furrowed brows and actually understood.

"It's basically providing code check-in and check-out services for the development team, that is, the service of putting code into the database and taking it out of the database."

"For example, like Google Chrome, the development team needs to constantly add new features and fix vulnerabilities. In this process, we can't directly edit and modify the running browser's source code, right? Because if the running source code has an error, it will cause the entire browser to crash, affecting the user experience. Therefore, we need a dedicated code management tool to solve this problem."

"This code management tool can help us build a code repository, which is like a large filing cabinet. It can be divided into different branches, that is, different versions of the code, according to different development needs, and saved separately."

"For example, we can copy the version that Google is currently running into this repository, then modify, delete, and test it. Once there are no problems, we can merge it with the currently running version. Users can then update or delete some code, and it becomes a new version."

"Wouldn't that mean every internet company would need to use it?" Hamilton exclaimed.

He hadn't realized this thing was so crucial; just like a computer needs an operating system, internet companies developing software also need it.

"That's right, every internet company needs it," Carl answered confidently. "The code management tool everyone uses now is CVS, which has a market share similar to Microsoft in the personal computer field, basically dominating the market."

"But if that's the case, is it still necessary for us to develop a new source code management tool?" Hamilton frowned again, asking in bewilderment, "Since CVS has already dominated the market like Windows, with a mature user base and a complete ecosystem, why would other companies abandon the CVS they're used to and adopt the new tool we're developing?"

Instead of answering the question directly, Carl Weller turned his gaze to Ernst, a hint of admiration in his eyes.

After a pause, he slowly said, "The reason we are confident in challenging CVS's market position is because Ernst proposed a disruptive concept: the distributed control concept."

Knowing Hamilton probably wouldn't understand, Carl once again took on the role of teacher: "CVS adopts a centralized control concept. Its core is a central server, where all the development team's code is stored."

"In other words, any internet company that wants to develop a new version of software or call code must connect to CVS's central server via the network, retrieve the code from the server, modify it, and then upload the code back to the server. This model is extremely dependent on network connectivity; once the network is interrupted, development work cannot proceed."

"Ernst's distributed concept completely breaks the constraints of this centralized approach. In a distributed architecture, every developer and every enterprise can build their own central server, or local code repository, on their local device."

"When developers make code modifications, they do not need to connect to a remote central server. Instead, they directly call the code from their local code repository and commit the changes to the local repository after completion. The central server only serves as a hub for collaboration and synchronization, essentially acting as a backup file repository."

"Every year, CVS causes code loss for many companies due to central server failures, network problems, and other reasons. For these companies, code loss not only means delays in development, but also requires additional funds to redevelop the code, which undoubtedly increases their cost burden."

"If we can develop a source code management tool based on the distributed concept and solve these industry pain points, do you think it will not be accepted by the market? It will inevitably disrupt the entire source code management tool market."

Hamilton understood and looked at Ernst. "So you want Google to be the guinea pig, to be the first to use and test this new version of the management tool once it's developed."

No wonder I called the three of them together today; it turns out there was another reason besides watching Microsoft's trial of the century.

Ernst smiled, which was taken as an admission.

In fact, he had many more ideas he didn't say; source code management tools were just the first step.

In addition to code management tools, he also wants to build a complete technology ecosystem, including a database management system.

Log monitoring tools, code editors, development and testing platforms, etc.

He wanted to use these basic tools to build a protective barrier, completely encircling Microsoft's Windows operating system and rendering it immobile.

As these basic tools become more widespread in the market, more and more developers and companies will get used to developing software within this ecosystem.

At that time, the operating system will no longer be the core hub of the software ecosystem, but will be completely reduced to a low-level operating platform.

>