Wei Qing: Step into the Unknown, What Opportunities Has Microsoft Seen in the Metaverse?

原创 精选
In this article, we invited Mr. Wei Qing, Chief Technology Officer of Microsoft China, to discuss the opportunities that the tech giant sees as we take a step into the unknown of the buzzword.

Technology is changing at an incredible pace, and there have been many exciting trends coming to our attention in the last few years. Therefore, it is not surprising that we have witnessed the emergence of many revolutionary technologies—the Internet, the Internet of Things, artificial intelligence, the blockchain, and, most recently, the metaverse.

In this article, we invited Mr. Wei Qing, Chief Technology Officer of Microsoft China, to discuss the opportunities that the tech giant sees as we take a step into the unknown of the buzzword.

How does the author of Snow Crash view the metaverse?

The term "metaverse" originates from the novel "Snow Crash", which celebrates its 30th anniversary this year. The author, Neal Stephenson, lives in Seattle and is a close friend of many Microsoft employees. Satya Nadella, the Microsoft CEO once stated that "Snow Crash" had a tremendous influence on the technology industry in his book book "Hit Refresh"; Kevin Scott, Microsoft's CTO, did an interview with Neal Stephenson last year and stated that Stephenson's humanistic approach to technology has profoundly influenced him.

Stephenson, who coined both the terms avatar and metaverse, gave an interesting answer to the sudden popularity of the buzzword: "They need a 'next thing,' you know, that they can use to drive their businesses forward, and so metaverse is kind of a catch-all term now for stuff that people want you to buy a few years from now."

(Image Source: Microsoft)

Furthermore, when it comes to artificial intelligence, Neal said: "AI will be called software when it's done." It seems that people are more concerned about the training stage of machine learning and that a lot of attention is given to post-trained models right now. It is, however, the actual application of a model in everyday life that measures its true success. Many modeling algorithms are very similar to the working of functions; you input the variables, and the program returns a result.

Identifying what is and is not a technical term is essential for business decision-makers or IT professionals. It is only after we have reached a consensus regarding "what it is" and "what it is not" that we will be able to work together to develop practical applications. 

This is a critical capability in this age of information explosion. Technology is made to be used, and truly superior technology is sophisticated and contributes significantly to the advancement of human society and individuals' wellbeing. In this regard, it will be necessary to discern whether it is "people-oriented" or "name-oriented"; whether it places equal importance on technological advancement and practicality. 

How should we proceed into the unknown?

In light of the social, economic, and technological changes that have taken place over the last few years, it could be claimed that we are in a time of extraordinary transformation that has sent the human species "into the unknown".

A significant part of the process is analyzing the changes and reconfirming the completeness of knowledge by using a rational mindset. This includes confirming that previous knowledge and cognition are still valid and updating those not. This can only be achieved if individuals think and judge rationally and accurately. Taking up this challenge requires conscious training for these capabilities, as it is crucial to our survival and development in the 21st century.

Our era presents both challenges and opportunities, which may position us to enter a new generation of technological advancement. However, as a starting point, it is prudent to adopt a humble perspective rather than waste time on definitions we cannot completely comprehend.

Information in books isn't always accurate. There is nothing wrong with saying that you are unfamiliar with something when you approach somewhere that no one has been before. Sometimes there might be no precedent to follow in the new zone, and for those who claim to have experience of the new continent, there remains a significant question mark.

The first step to learning is admitting that you don't know something. As there are so many unexplored theories and concepts, it is even more crucial to know that we cannot drive on the appropriate pathway simply by studying or imagining a solution without any practical cases. When dealing with new things, especially those far from consensus, what is vital is one's practice and the insights gained from it. The "learning by doing, doing by learning" method, that is, a rational way, may help us find a clear direction to the future by applying insights derived from personal practice to form a correct judgment and then verifying and correcting those judgments in continuous actions.

People tend to make fragmented conclusions when coming up with new ideas, especially when they are unfamiliar. Here, it is critical to think logically to avoid assuming that new terms mean new concepts or that big numbers are synonymous with superior technology.

The "fourth industrial revolution" may have taken place in certain places, but it does not mean that all parts of the world are experiencing the same development. In addition, when we examine the characteristics of the times, we can see that the focal point of society is shifting from "substance" to "energy" and then to "information", which corresponds to the transition from an "agricultural society" to an "industrial society" and then to an "information society".

The current dissemination of information and its overuse have a significant impact on people, physically and psychologically. In comparison with the Renaissance and Enlightenment periods in history, it appears that we are seeing another round of social changes.

Therefore, we must adopt a philosophy of evidence and adopt a systematic approach to admit our ignorance and be willing to apply it in practical ways to unite all our efforts in a practical direction.

Besides, there are no eternal leaders. Satya Nadella frequently reminds employees to distinguish between "leaders" and "incumbents" in the market. It should be pointed out that the incumbents mentioned are the same ones discussed in Clayton Christensen's book "The Innovator's Dilemma", where he uses it as a corresponding term to market challengers.  

As an incumbent, you realize that your successes are not eternal and should not let pride get in the way of your actions. A challenger could appear at any time and break the mold. As for individuals who have not yet achieved the status of a leader, they will have the courage to study and practice, as they understand that lagging is only a temporary status. This is Nadella's reminder that both success and failure are dynamic and that both are necessary on the way to building a better future.

Lastly, it would help if you did not look for partial optimal solutions too early in the process. Human-machine interaction is an eternal topic, and since we have entered the new era of "human-machine" interaction, we can learn something from machines.

The initialization of a neural network's parameters is the first step toward machine learning after setting up the structure, which essentially requires some probability distribution.

Thus, we should not strive for an optimal solution too soon but instead continuously improve our learning until we find a relatively optimal one.

Having the attitude of "admitting mistakes, understanding mistakes, and correcting mistakes" is a helpful way to tackle uncertainty in a new industry, no matter whether you are dealing with machine learning or human civilization. It is inevitable to make mistakes in a new field, but we should make high-quality mistakes, that is, mistakes from which one can gain insights. The era of uncertainty requires us to concentrate on "understanding" what is wrong and "correcting" it as soon as possible.

It is precisely at this point where the capabilities of machines can be of great benefit by using them to identify what is wrong as quickly as possible and make the necessary corrections in a short time. This ability to enable humans to "detect" and "correct" mistakes has emerged as the main advantage of digital technology for human wellbeing. This is also the digital feedback chain proposed by Microsoft based on the principles of cybernetics.

By examining the concept of "metaverse" within this new "human-machine" relationship, we can better comprehend its value and limitations. 

Metaverse or digital twin? It doesn’t matter.

As soon as we grasp that the relationship of nouns and their essences are just signifier and signified, we no longer have to struggle with questions like what is the difference between a digital twin and a metaverse.

It does not matter whether we call it—the metaverse, digital twin, cyber-physical system, Semantic Web, or Web 3.0—because all the topics deal with a timeless issue in human development: how we use tools to enhance capabilities.

Tools have been used to increase human strength in the past, but now they are utilized to expand our thinking abilities. The computational power of machines has enabled the mapping of the physical world that humans live in into the tensor space of machines or cyberspace.  

Human visual and auditory representations simulated by software and hardware capabilities, coupled with tactile, olfactory, and gustatory representations that are gradually being realized, will model, map, simulate, compute, optimize, and predict the physical world. These capabilities will contribute to creating what Jean Baudrillard called a "hyperreal simulacra".  

Though this simulacra is still very crude technologically, the potential that it has and the effects it may have on human thought call for careful differentiation and definition of the physical world versus cyberspace.

(Image Source: Microsoft)

There is a tendency nowadays to overemphasize the value of cyberspace. This is necessary for its continued existence, but isn't the actual experience of physical space more valuable for humans? This is why it is wise to look at it in a balanced way. As a result of a holistic analysis, ubiquitous computing power is utilized to support two parallel worlds. Objects in the physical world are mapped into the virtual space after simulation, calculation, and prediction, and then the results are applied to reinforce human capabilities in the physical world.  

Considering this, isn't the navigation application we use every day without even realizing it a prime example of a metaverse application that has already made significant contributions to human society?

Imagine numerous cars and human behaviors being simulated in cyberspace and subsequently computed, simulated, and predicted. Later, route suggestions will be provided to enable human or autonomous driving. Depending on the concept's prerequisites and definitions, we can probably define it as a metaverse or something similar.

Let's take the current state of technology maturity into account rather than focusing on the virtual space experience, which simulates human vision and hearing. We see real physical space and virtual cyberspace mapping each other's applications on the ground. This has already started to move toward the most basic human needs, such as food and clothing supply, and solving these problems should be the primary goal of technology.

Computing power in cyberspace is not infinite, while at the same time taking up physical resources, time, and capacity. This valuable computing resource should prioritize addressing the Sustainable Development Goals defined by the United Nations that are focused on the survival and wellbeing of human beings. When the basic needs are met, such as food and clothing, education, and medical care, the meaning of the metaverse in the physical and virtual world may become more clear to the public.

(Image Source: Microsoft)

In this way, Microsoft Azure's vision of "ubiquitous computing and ubiquitous intelligence" and its human-centered approach will become increasingly proactive and valuable to society. As a result, you will not have to hassle yourself with terms like cloud computing, edge computing, etc. According to Nadella's definition of Microsoft Azure when he first became CEO, it is an operating system for the cloud. Operating systems are made up of computing, storage, networking components, and a user interface is added on top of that. The principle remains the same, whether in cloud-native computing or the very hot metaverse. The driving force behind it is the need for people, not technology.

Azure could leverage its global computing, storage, and network resources to make the computing power of cyberspace work better for humans. Therefore, combined with the ability to present mixed reality enterprises that can embed intelligent digital feedback chains throughout their products, employees, customers, and operations. We call this an end-to-end metaverse solution.

(Image Source: Microsoft)

Application opportunities in the metaverse

Now let's return to the fundamental concept of growing "human-machine" interaction. In that case, we may take a broader look at the potential of integrating physical and virtual space, or a "metaverse opportunity cube" as the image shows, which has 27 "opportunity points" in three dimensions plus a timeline.

(Image Source: Microsoft)

Horizontally, "targets" include consumers, enterprises, and industries. For functions or content task objects, it has content creators, platforms, and users who could work in physical, cyber-physical, and pure cyberspace. There are, of course, more extensive legal, administrative, technical, and standard roles, but they are currently omitted from the description of this cube.

Historically, there was much discussion about digital twins and gaming applications. What is happening now is similar, only integrating virtual reality more closely. In the coming years, application capabilities and experiences will continue to evolve, deepen, and improve.

A consensus has not yet been reached regarding the definition of the metaverse, and the concept is still in its infancy. From Microsoft partner DataMesh, here is another definition of a metaverse application. By going the other direction, they call their product "FactVerse," which helps many companies use the virtual environment's computing power and simulation capabilities to enhance the physical space, utilizing a digital twin to make the real world "visualized-computable-operational." These applications, whether virtual or physical, should benefit individuals.

(Image Source: DataMesh)

Microsoft's digital twin solutions for the industrial sector are also an example. In addition to content producing and consuming, there are platforms of transmission and sharing, which is an integral component of the digital twin in the industrial sector.  Why is this not a metaverse?

(Image Source: Microsoft)

Microsoft's Azure Digital Twins additionally integrate the physical properties, relationships, and action logic of the objects with the computational, simulation, and machine learning capabilities of the Azure virtual environment via a Digital Twin Definition Language (DTDL) that is compatible with W3C's RDF, OWL, and SHACL.

Based on the robust development and Azure DevOps environment, we can build a virtual world that can be extended and enhanced by a digital twin. Furthermore, it can also be assumed that, depending on how the metaverse is created, as well as any limitations that may affect its realization, we might already be dealing with a few applications and have stepped into that unknown metaverse for a certain time.

责任编辑:庞桂玉 来源: 51CTO

2022-08-31 15:13:11


2022-08-31 14:39:45


2022-08-30 19:50:34


2022-08-30 22:45:36


2021-06-22 06:24:57

Linkerd Ingress 流量网络技术

2011-04-19 14:02:09


2022-08-30 19:41:09


2010-09-08 09:41:03


2022-08-31 08:45:47


2009-04-22 17:18:29

Vxworks驱动加载step by ste

2021-06-16 17:42:48

Linkerd 配置CPU

2021-06-22 06:16:24

Linkerd books webapp

2021-12-23 15:11:46

Web 3.0元宇宙Metaverse

2021-06-22 06:41:38

Linkerd 安装多集群组件网络技术

2021-06-17 06:13:29

Linkerd Prometheus 网络技术

2022-08-31 10:53:46

AIAI chatbotmetaverse

2022-07-08 00:08:48


2020-12-17 16:53:23


2021-06-17 14:29:39

Linkerd 分布式跟踪Linkerd 2.1

2021-06-17 06:20:43

Linkerd Kustomize网络技术