news

한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina

compiled | chen junda

editor | panken

zhidongxi reported on september 27 that recently, before and after the meta connect 2024 conference, meta founder and ceo mark zuckerberg accepted three in-depth interviews. these three interviews were full of useful information. full and full of exciting stuff. xiao zha took the initiative to talk to nvidia founder and ceo jensen huang "close to",againattack appleproduct development model and closed platform, and also admitted that he had committed the past 20 years2 big mistakes, he also revealed that orion, the next generation of more cost-effective holographic ar glasses, willavailable to consumers。

these three exclusive interviews were hosted by acquired podcast, technology media the verge and youtube technology blogger celo abram, with a total length of more than 3 hours.

zuckerberg said huang was "very candid" in sharing that he would "never start a business again". the process of starting a business was extremely painful, as if he was always on the verge of life and death. in the early days of entrepreneurship, when yahoo tried to acquire facebook, zuckerberg became the only executive who opposed the acquisition. in the end, he was alienated from his family and almost all the top executives left the company.

in the interview, xiao zha directly characterized apple as meta’s “main competitors” and called the competition between the two companies the development concept of open and closed source.ideological conflict. he also called meta the opposite of apple. meta likes to iterate quickly, but this does not prevent them from launching excellent products.

meta has been deeply involved in the field of holographic display for 10 years. xiao zha revealed that theyorion alone cost us$5 billion (approximately 35.1 billion yuan),but i was not completely satisfied with the first-generation product, so i changed my original plan to sell it as a to-c product. in the future, meta's smart glasses product line will cover three price ranges: no display, heads-up display and holographic display.

open source is the key to zuckerberg's success. meta's open source in computing infrastructure, llama model, etc. allows them to enjoy the scale advantage of the supply chain and significantly reduce costs. zuckerberg put aside his rhetoric and wanted the open source platform to become the winner in the battle for the next generation platform in the next 10-15 years.

looking back on his entrepreneurial journey, xiao zha also admitted that he had committed a "1 and a half years of mistakes” and is currently working on a “20 years of mistakes"atonement."

in 2012, before and after facebook's ipo, they decided to use html5 as a universal platform, aiming to use a set of rules to open up the mobile platform that was flourishing at the time. however, this decision greatly affected the user experience and caused the stock price to halve. at that time, facebook had to suspend the development of new features for a year and a half and rewrite its own programs.

in 2016, facebook was criticized for its controversial role in the us election. zuckerberg said that both the left and the right were dissatisfied with his decisions. this series of wrong decisions caused serious damage to facebook's reputation. zuckerberg believes thata political miscalculation could be a mistake that will affect them for 20 years,he emphasized in another interview this year that meta's platforms would reduce the push of political content during the current us election.

in recent years, many people have noticed thatxiao zha seems to have changed from a "robot" to a "normal person" with flesh and blood.. zuckerberg shared in the interview that he made a statue of his wife and designed t-shirts with famous designers, so he no longer always wears the same clothes. he started a cattle farm in an attempt to raise the best beef in the world and was involved in some scientific research.

▲zuckerberg was interviewed wearing a t-shirt he participated in designing (source: acquired)

talking about these changes, he believes that the epidemic has given him the opportunity to step away from work and reflect, and now he hopes to "do inspiring things with inspiring people." he also said before the interview officially started that he "will never apologize for anything anymore."

the following is a compilation of parts of three recent interviews with zuckerberg.(in order to improve readability, zhixixi adjusted the order of some questions and answers, and made certain additions, deletions and modifications without violating the original meaning. the interview contents not specifically marked are all from acquired):

1. huang renxun’s point of view is very reasonable. entrepreneurs are always on the verge of life and death.

david rosenthal: if you had all the knowledge you have now when you started a business, would you still choose to start facebook?

zuckerberg:i think huang renxun’s point of view is very reasonable: the entrepreneurial journey is extremely challenging, especially in the early stages of entrepreneurship.entrepreneurs always have the feeling of being on the verge of life and death., facing strong uncertainty.

▲zuckerberg and huang jenxun have frequently interacted recently (source: instagram)

when i look back on the experience of starting a business, there are always some good memories, but starting a business is definitely not the most pleasant experience in my life, and i definitely don’t want to go back to the past and experience the feeling of starting a business again.

lao huang said before that if he went back to the past, he would definitely not start a business again. i think this statement is very honest, and i felt the same way when i first heard it. for many people, if they had known how hard it was to start a business, they might not have embarked on the journey.

human nature has an "advantage";that we always underestimate how painful things will be, which allows us to do some very meaningful things.

ben gilbert: what does learning from suffering mean to you?

zuckerberg:i think that constantly running into walls can help you understand what is really important to you, and it can also give you a clear positioning of yourself. that’s really the entrepreneurial journey and the beauty of building something new.

when people start a business, they often write down what the company's values ​​are. but in fact, values ​​are not slogans written on the wall, but need to be practiced in real life. only when you have to make a choice and face a challenge will you understand what you really care about and what is important.

2. meta will promote the open source platform to win, and apple is its main competitor.

david rosenthal: at this point, i feel like you have accelerated your "annual challenge." when both of us (interviewer) were kids, i would be inspired by your annual challenges and thought they were some cool stuff.

then we all grow up and have our own companies. i feel like i don’t have so much time, but you seem to have more time to do these challenges. now i feel like you're doing weekly challenges, you're designing t-shirts, making sculptures, raising cattle, etc., i feel like you have a challenge every week.

zuckerberg:i'm trying to do something inspiring. i'm also very competitive.

david rosenthal: who are your competitors?

zuckerberg:do i have any personal competition? i just wanted to do more and got involved in more extreme sports, boxing and stuff like that. as for the company, we face competition on many fronts, both in our social media business and on our platforms.

i think apple is a bigger competitor than people think. they may think they are doing something different than us, but i think in 10-15 years,this ideological battle over next-generation platform architecture will intensify, will they be the closed, integrated model that apple has always been?

i think different times have different ways. computers of different generations, including pcs, mobile devices, etc., have closed, integrated versions and relatively open versions. i think everyone may be affected by a lot of "recency bias" at the moment, because at this stage the iphone is basically winning. although there are more android phones, apple is ahead in terms of intellectual property rights.

almost everyone here has an iphone, which lends itself to a perception of superiority in the way they do things, but i don't take that for granted. in the pc era, the open ecosystem of windows is the ultimate winner.my goal for the next 10 to 15 years is to build the next generation of open platforms and let open platforms win, i think this will make the technology industry more dynamic.

closed and integrated models do have advantages, and i think apple will definitely still have a place,i expect they will be our main competitors.

this is not just product competition, this is to some extent a very deep value-driven and ideological competition, involving what the future technology industry should look like, whether these platforms should be open to developers, such as llama and ai or glasses. , should people like me who started a business in my dorm room be allowed to build the next generation of great things without having to ask for permission.

3. meta is the opposite of apple. rapid iteration and great products do not conflict.

ben gilbert: in the research i conducted prior to our interview, someone described you as a “strategist.” they say you play business as a turn-based strategy game, and the secret to your victory is to make sure you get more turns than everyone else and to make sure you learn more from each turn than the other players. , do you think this accurately summarizes meta's situation?

zuckerberg:i do like the expression "turn-based strategy". in my opinion,great engineering products and rapid iteration are not mutually exclusive.there are a lot of great companies that want to build super high-quality products and have great capabilities. but it's also a strategy to launch something less complete quickly.

i'm not saying that our strategy in this regard is the only effective approach;but i think we are actually very much the opposite of apple. their product is also pretty good, but it takes a lot of time to polish. this approach may be effective for their business or more consistent with their culture. but for us,we would quickly release products that were less than polished, sometimes products that we ourselves would be embarrassed to look back on.

we prefer to publish as quickly as possible to get feedback. we need to test various reasonable hypotheses. sometimes we find during testing that our assumptions don't actually make sense, but most of the time we find that we can significantly improve the product just by continuing to work on it for a few months.

i think we should promote a culture that values ​​launching products and getting feedback, rather than expecting to always get rave reviews every time we launch.if you keep immersing yourself in product development and always think about launching a product with rave reviews in one fell swoop, you will waste a lot of time. the company could have used this time to learn a lot of useful knowledge and integrate it into the next generation of products.

of course, we won't launch products that we think are bad, but we will also make sure we launch them as quickly as possible so we can get feedback and see what their most common uses are. for example, although we now know that ai can bring revolutionary development to many things, we are not so clear about what the initial use cases of these valuable technologies should be.

if we release something that's really bad, we won't be able to collect valid data because the product will have no use at all. but i do think we should roll it out early and see if our previous assumptions are reasonable.

4. holographic ar glasses have been developed for 10 years and can become a perfect ai assistant.

david rosenthal: speaking of challenges, i want to talk about some of the challenges in your entrepreneurial journey. by our estimate, you may have faced more significant challenges in the past 20 years than any other major company in the past. tell us about those challenges.

zuckerberg:meta has been building social experiences for the past 20 years. it started as a website and then a mobile app, but i never saw ourselves as a "social media company."

we are not a social media company;we are a company that builds social connections. we build the future of human social interaction., will not be limited by the small screen of the mobile phone.

when we started our business, we were just a bunch of kids. we didn’t have the resources or the time to define what the next generation computing platform should look like. in addition, facebook was founded around the same time as new platforms such as smartphones emerged, so we didn't have the opportunity to play a role in the development of those platforms.

in the next phase, our company’s theme is to build the ideal social experience, rather than developing within the scope of what others' platforms allow. we need to start from first principles to analyze what is an ideal social experience. i think what people want may not be a cell phone, because when we look down at our phones, we also lose focus on the people and things around us.

i think the ideal form would be glasses. on the one hand, smart glasses can see what you see and hear what you hear;can become the perfect ai assistant for users, because they understand what you're doing.

▲zuckerberg wears ray-ban glasses developed by meta (source: meta)

on the other hand, these glasses can project images, similar to holograms. this way, your social experience isn’t limited to limited interactions on your phone screen. we can even have “offline” conversations like today without each other being present.

another problem is how to create a real "sense of presence". the physical perception of another person's presence is a very magical experience, and humans are actually very physiological animals. we like to attribute everything to our minds, but many experiences are deeply bound to the body.

we can achieve this physical "sense of presence" through technologies such as smart glasses and holographic projection., and it doesn’t take us away from what we are doing. the real world and this social experience are deeply integrated. i think this will become the ultimate digital social experience and the ultimate embodiment of ai. we may be able to communicate with someone’s ai avatar through smart glasses.

this vision is an ambitious project, how do we achieve it? we've actually been working on this for 10 years, with many different challenges to solve. first, a new stack of display devices needs to be built. we need more than just the kind of screens we have in our phones. that kind of screen has a long history and is connected with televisions, computer monitors and other equipment, and the supply chain has also been well optimized.

the display device stack associated with holographic displays essentially needs to be developed from scratch and needs to be miniaturized so it can be stuffed into devices like glasses. you also need to install chips, microphones, speakers, cameras, eye tracking and other equipment in the glasses so that the device can understand what you are doing. it also needs a good enough battery to ensure a day of battery life.

we have cooperated with the glasses manufacturing giant essilor luxottica, the manufacturer of ray-ban glasses, to explore how much technology we can currently integrate into the form of glasses and what uses it can play. when we first started working together, it was like an experimental practical project, exploring the possibilities of augmented reality (ar) that might be realized in the future.

5. the development cost of orion reaches 5 billion us dollars, and the next generation product will be reduced in price and fully released.

the verge interview content:

zuckerberg:what i see in the orion holographic ar glasses is the prototype of a next-generation computing platform that will be used by billions of people.

the verge: orion was originally a consumer product, right?

zuckerberg:yes,originally orion was supposed to be our first holographic ar glasses for consumers., but we're not sure it will be successful. this product did exceed our original expectations, but our own expectations were not very high, and this product has not yet met all of our internal company requirements.

therefore, before officially launching it as a product, we hope to make orion smaller, brighter, higher resolution, and more affordable. i think we had expectedsecond generation orionit may become the first generation of consumer-oriented holographic ar glasses we launch. we will use the current orion to optimize the software experience with developers, so that when we are ready to launch the next generation of products, the products will be even better.

around 2022, i decided to reposition orion as a developer kit and not for consumers. i actually prefer to launch the product as soon as possible, but i think that decision was the right one. the product needed to meet a number of strict requirements, especially in terms of appearance. my current glasses are pretty good.weight does not exceed 100g, but we hope to launch better products.

the current orion is already a good pair of glasses, but i wish they could be smaller so they look more fashionable. our experience with ray-ban glasses is this: people don’t compromise on fashion. i think the reason why people like ray-ban glasses is not just because of their function, but also because the glasses themselves look good. this is also the goal for future orion glasses, and currently it has not reached my goal.

in terms of price, orion will definitely be more expensive because it uses more technology. however, we still hope to set its price within the range acceptable to consumers. the current orion is well beyond that range, and i want to release it fully when the price drops to a level that is acceptable to consumers.

the verge: so you’re going to skip selling developer kits to enterprises? apple and snap have taken similar steps before.

zuckerberg:we do use it as a developer kit, but we mostly use it internally and with a few partners.

at present, meta has a leading position in the world in the field of ar/vr/mr. we already have a large number of talents internally and have established good cooperative relationships with other companies. we don’t need to launch a developer kit, it’s enough for us to develop it together with our partners.

the verge: there have been many reports about your spending on reality labs, so it may not be possible to give a definite figure. however, can you estimate how much it cost you to develop orion holographic ar glasses in the past 10 years alone? did it exceed us$5 billion?

zuckerberg:yes, almost. many people think that the bulk of reality labs’ spending is on xr headsets, but we have publicly shared before that the budget for the glasses project is larger than both vr and xr.

6. in the future, product lines will remain diversified. smart glasses will have three product lines.

celo abram interview content:

zuckerberg:in the process of developing holographic ar glasses, we have also developed other types of products to help us achieve our ultimate goal. we built glasses without a display, and along the way we learned how to pack as much technology as possible into a pair of stylish eyes. this product is the glasses we developed with ray-ban.

at first, we thought this product was just a learning phase for us, but we now understand that we can significantly reduce the cost of this product and then make it a permanent product line. this type of glasses will get better and better, especially suitable for ai scenarios. although there is no display, we can still talk to it and get responses.

in addition, in addition to holographic ar glasses, we will also launch mid-range products with a viewing angle of 20-30 degrees.this head-up display still has great value, which allows users to see content, send text messages and display text, and more. this product may be more expensive than a display-less version, but it will be cheaper than holographic ar glasses.

additionally, i think the original xr headsets are here to stay. because no matter how much progress we can make in miniaturization technology, we will still only be able to pack full computing power into the head-mounted display device.

our mission is not to make advanced products that only a few people can use;but through innovation, everyone can use these products. we just released the new xr headset quest 3s, which provides a high-quality xr experience at a price of $299. when we launched the $500 quest 3 last year, i was already very proud. this is the first high-quality, high-resolution, color mr device at a fraction of the cost of our competitors. now we have further enhanced this product line.

i thinkthe above products will become important long-term product lines, including no-display glasses, heads-up display glasses, holographic ar glasses and mr headsets.

7. from the beginning of starting a business, i believed that facebook was more than just social media, and it was necessary to have a strong technical foundation.

ben gilbert: let's be clear, is this what you think facebook should be?

zuckerberg:yes, indeed. i had similar thoughts from the beginning.

i started facebook when i was in college, and then several of the founders came to silicon valley because we believed that this was where all startups came from. when we got off the plane and drove down highway 101, we saw ebay, yahoo, and other great companies, and we thought, maybe one day we will build a company like this.

we had already started the facebook project, but i felt that facebook itself was not enough to become a company.i don't think our future will just stop here. after ushering in new platforms such as ar glasses and headsets, we have witnessed the huge changes brought about by ai.

ben gilbert: i want to make it clear to the listeners that you guys were shipping this product before big language models came along, or at least not before the public realized that chatgpt was coming. these products are not defined as ai devices during the manufacturing and sales process. this only emerged after the listing.

zuckerberg:yes, my prediction a few years ago was that ar holographic images would be implemented before the emergence of powerful ai. now it seems that the order is reversed. when the ai ​​craze came, we also realized that the glasses themselves are actually a good product, equipped with cameras, microphones and speakers, and people can interact with them in the form of voice.

i remember contacting alex himel, who was running this product group, and telling him that we should pivot and make meta ai a primary feature of our devices. the next week they developed a prototype that we all thought could become a successful product.

ben gilbert: thank you for sharing. i guess my question is why meta has performed so well through multiple iterations and multiple technology waves to become one of the most valuable companies in the world. there are many people who believe that what facebook is doing is outdated and will fade away. yet you are still very active. what is that dna message that runs through facebook and meta?

zuckerberg:i think of us as a technology company focused on human relationships rather than specific types of applications. we have never thought of ourselves as a website or a social network. for me, creating glasses that allow people to feel like they're with another person no matter where they are is a natural continuation of our app.

success depends on how you define yourself and how you structure your company and equip it with the capabilities to complete these projects. i think it is necessary to build the company into a strong technology company, and many companies define themselves too narrowly.meta is currently able to do these things well because we have a very strong technical foundation.

8. meta will cease to exist without open source, and the open source path is suitable for its current market position.

david rosenthal: many people believe that meta is the biggest "beneficiary" of open source technology in the modern world. do you agree with this view? can you talk about your relationship with open source?

zuckerberg:i think almost all big tech companies are primarily using open source technology stacks right now. i don't think we could have built it without open source. i feel like this is true of any new company since the late 90s. for us, open source is very important and valuable.

david rosenthal: you are the first big company based on the lamp stack.

zuckerberg:yes, this allows us to develop and iterate quickly. but our relationship with the lamp stack is also interesting. because we were founded later than google, and google was the first company to build this kind of distributed computing infrastructure. initially, they chose to keep the technology proprietary to gain a competitive advantage.

then we realized we needed these technologies too. we developed it ourselves, but ultimately decided to open source it because google already owned it and we had no competitive advantage. by open source, we attract a community that builds around it. this technology doesn't directly help us compete with google.

but through open source initiatives like open compute, we shape industry standards. now all cloud service platforms basically use open compute, so the supply chain is standardized around our design, resulting in a significant increase in supply and a significant reduction in production costs. we've saved billions of dollars and the quality of the products we use has improved, so it's a win-win.

▲open compute project official website (source: open compute)

but to make this all work, we do a lot of open source stuff and a lot of closed source projects. i'm not an extreme open source person. i think open source is very valuable, but i also think open source is very suitable for our current position in the market, and the same is true for ai, such as the llama project.

we want to make sure we have access to leading ai models, just like we want to build the hardware to create the best social experiences over the next 20 years. we have experienced too many frustrations in relying on other platforms,we're a big enough company now that we don't have to rely on anyone.

we can build our own core technology platform, whether it's ar glasses, vr or ai, so i think this is an important goal for us. but these things are not single pieces of software, they are an ecosystem. they only get better when others use them.

for us, this has great benefits and fits our philosophy very well. we've also experienced a lot of platform-side blocking of development, especially when trying to develop on mobile platforms, which can be really frustrating.

9. the political environment has undergone tremendous changes. xiao zha said that he had made a serious misjudgment.

ben gilbert: you've accomplished a lot, but you've also experienced a lot of criticism. if you were asked to evaluate this company, what you've created, and the criticism it's received over the years, which criticisms do you think are justified and why?

zuckerberg:we messed up a lot, and there was a lot of legitimate criticism. one of the things i've reflected on over the past decade or so is that the political environment has changed dramatically. before 2016, except for the ipo period, our company's reputation was basically relatively positive.

then after 2016, after the election, basically for a period of time, the company's reputation was basically negative. i think we need to have a correct understanding of where the company is in the world and in history. we rightly define ourselves as a technology company, and we rightly define ourselves as a company that creates human connection.

i didn't have much knowledge about the political environment, so i misdiagnosed the problem.i don’t want to oversimplify this, but we did a lot of things wrong and we did a lot of things right. looking back, one of the things i regret is that we accepted other people's opinions about some of the things we did, that they thought we were wrong or responsible.

i don't think we really made a big mistake politically.we did mess up a lot of things that we need to fix and improve. but there is a view that when someone points out a problem in the company, the company's instinct should be to take responsibility, and even if it is not entirely the company's responsibility, it will take responsibility for it and solve the problem.

but when faced with a political problem, some people are pragmatic and want to solve the problem, but some people are just looking for someone to blame. if companies continue to take all the responsibility on themselves and view the problem as a company problem rather than the political environment as a whole, people will start to blame social media and technology companies for all the problems in society. superior.

if we promise that we will truly do our best to solve these problems, some people will take this to add insult to injury. to be honest, i think we should have been more firm and clear about which issues that arose at the time that we were actually involved in and which issues we were not.

my guess is that if the ipo-related decisions were a mistake that lasted a year and a half,political miscalculation could be a mistake that will affect us for 20 years, which started in 2016. we've been working very hard to address a lot of issues and figure out the right tone for dealing with tricky political dynamics.

i think we've found our footing in terms of principles, areas where we think we need to improve, but when people make accusations about the impact of the tech industry or our company that have no basis in fact, i think we should be more resolute fight back.

i think it's going to be another 10 years or so before we fully complete the cycle and get our brand back to where i didn't screw it up. but from a macro perspective, 20 years is not that bad. we will overcome the difficulties and we will become stronger.

we appeal to both sides of the political spectrum, with some feeling we don't take enough responsibility and some feeling we interfere too much. it takes us a long time to get out of this.

ben gilbert: do you have a reasonable solution framework now?

zuckerberg:i think a lot of the issues have been studied. let me just say that we have learned that we should try to support more scholars to conduct research in related fields in advance. because when we've already been accused, credibility has been damaged. if we can let third-party scholars conduct research in advance, we may be able to find that social media and various problems in society may not be necessarily connected.

10. due to the yahoo acquisition, a same-share-different-vote structure was adopted. the most important thing for starting a business is to remain flexible.

david rosenthal: facebook's organizational structure and operating methods were very unique when it was listed. you owned super voting shares, and the company adopted a structure of different rights for the same shares. why are these so important to you?

zuckerberg:in 2006, yahoo wanted to buy the company for $1 billion, everyone on our management team wanted to sell it, the board tried to fire me, and basically everyone else on the management team left over the next year. yes, because i did not communicate well. i don’t blame them, i didn’t communicate the long-term vision well and didn’t think through it at the time.

i wasn't thinking about this from a company perspective at the time. i just think it's a great project that's probably going to be around for a while and i think it's going to be a big deal. but don’t know how to think in terms of long-term financial planning.

when yahoo made an acquisition offer, it was an entrepreneurial dream come true for many people. it seemed like i had to accept everyone's proposal because i didn't have the ability to articulate the long-term direction of the company at the time, and they might not have the confidence to continue with the project. after this incident,i think i need to establish a new governance structure to try to avoid the previous situation.i can also be considered as "learning from pain".

ben gilbert: and you started generating cash flow very early on, and the company has been operating steadily, and you can continue to start new projects without selling.

in the 21st century, many people choose this path because they think entrepreneurship is glorious. but you don’t seem to enjoy being a start-up, but instead want to transition into a stable business as quickly as possible. do you have any advice for founders who dream of starting a business and kind of romanticize it?

zuckerberg:starting a business isn't bad, but some people start a business because they want to delve deeper into an idea, and i think that model is a little dangerous.

entrepreneurship requires being able to be flexible and pivot until you figure out what works. i didn't think facebook was going to be that successful company at first because i was working on 12 other different projects in school.

i think it helps to be flexible.once you hire a group of people, it's much easier when you can have a meeting in your own head to decide the direction you want to go, and a lot less egotistical and opinionated when you decide to change direction. that's why we try to make the company as lean as possible.

we need to strike a balance, and in the early stages,on the one hand, we should not give the team too few resources, nor should we give too many resources at once., causing them to be unwilling to change and innovate. without change and innovation, it will be difficult to attract outstanding talents.

11. platform taxes cost meta half of its profits. we firmly believe that ar and smart glasses have huge potential.

ben gilbert: but you're spending an astronomical amount of money on reality labs, and you're not making that much money. of course i'm asking this a bit arrogantly, but i want to know why you put reality labs in another category?

zuckerberg:i think when you get to the size of our company now, sometimes you just have to think about what do we want to do in the next 10 to 20 years, what do we think will be important? we talked about making our own luck, and i think we have a rough idea of ​​where things will go.

i’m pretty sure glasses, holograms, and ar will become ubiquitous products. it's like everyone used mobile phones before, but then they all switched to smartphones, and then more people had smartphones.

if we just had everyone in the world who already wears glasses upgrade to ai glasses, this would already be one of the most successful products in history, and i know this product has even greater potential.

there is also the need to control our own destiny, which is strategically valuable. we once estimated that the various taxes and fees levied on us by the platform cost us a lot of money on our core application family. they'll tell us we can't run the advertising business the way we want, or we can't launch certain products.

i think if we had our own platform, our profits could probably double.from a purely monetary perspective, this is also valuable.

i've also learned a few things since the yahoo acquisition days. if i didn't control the company, i might not be able to convince every investor to invest in reality labs to the extent that we do now, but at least i can articulate a rationale for why i believe this will be good in the long run.

12. it is not difficult to go from failure to success. what is difficult is to endure the pain of transformation.

david rosenthal: let's switch gears a bit, can you take us back to the eve of facebook's ipo? i want to ask you what you were thinking when you decided to launch facebook ipo and adopt html5 as mobile technology?

at the time of ipo, facebook's market value reached us$100 billion, but it shrank by 50% in the next three months. this may be related to your decision. did this incident have any impact on the way you deal with ai technology?

zuckerberg:i think this is a completely different technical issue. our tradition is to build websites based on web pages, and we are used to building one thing and being able to continuously deploy it, which fits our iterative style.

around 2012, this app model suddenly emerged, where we had to build a different app for each phone, and it had to be reviewed before we could release it, and we had to wait weeks.

we thought this was too bad and came up with a new idea. we wanted to develop a web-based platform that could update our app on a daily basis and update it across all mobile platforms including android, iphone, blackberry and windows in one release. we felt we could offset the disadvantage of not developing native products for every platform through the speed of iteration and the advantages of focusing on one platform.

it turns out we were wrong,having native integrations is crucial to providing a good interactive experience.there was a time when we had to write our app from scratch, which also coincided with the rapid growth of mobile devices. we weren't generating any revenue from mobile devices because on desktop we could benefit from sidebar ads, but on mobile we needed to figure out how to integrate ads into the user experience.

ben gilbert: there was no such thing as in-feed advertising.

zuckerberg:yeah, that's some of the stuff that our team does. advertisers still like to use specific formats, but we tell them that in the future your ads will look like a message in the information stream, which is a big challenge for advertisers.

for people, the original information flow is the most important part of the product, but now we have to add advertising, which is also a challenge for people who use the product. we need to make the application better, and we may spend a year or more working on this problem.we basically took a break from developing new features because rewriting is inherently difficult.

if you look at the history of the technology industry, there are many examples of this, such as the netscape browser and other attempts to reinvent the wheel. they needed to rebuild their technology platform and also try to add functionality. but changing the underlying platform carries a huge risk, and you're likely to fail.

we feel it's important to minimize the chance of this happening, so we won't be releasing any new features, just rewriting existing features to make them faster. but when we do that, the percentage of traffic we can monetize is shrinking because web users are declining and mobile is growing.

david rosenthal: the web version was your only viable business model at the time.

ben gilbert: after going public, you will also need to conduct financial reports every quarter.

zuckerberg:but the thing is, we actually have a pretty good idea of ​​what we need to do. i think strategically, it's actually harder to know what to do when you're winning. when things are going well, it’s important to look at how you can move from victory to greater victory. but when you lose, usually you know exactly what you have to do. i think the key in this situation is whether you have enough ability to endure the pain of transformation.

at that time, many people in the team believed that since we were a listed company, investors would definitely be unhappy if we did not make money for a year and a half. but a year and a half is a short time in the grand scheme of things. this year and a half was a very painful experience, but we finally came out of it and developed pretty well.

i think internally the company realized we were improving earlier than the outside world. employees know we're doing the right thing, and they know we're executing responsibly and with focus. of course, it was certainly not a fun experience looking back.

i think what is more difficult than the above adjustments is to change a mechanism that was originally working well. this is very challenging for the company culture.

13. some companies do not deserve to call themselves “technology companies”. rapid iteration and curiosity are the secrets to victory.

david rosenthal: can you take us back in time and tell us a story about that era? at that time, there was also friendster, a social media company like myspace. how did you win?

zuckerberg:i think this is a combination of product issues and technical issues. we can define the product very broadly, but then the related technologies will be difficult to solve. the company needs to position its products correctly, but it also needs to be the strongest and best in relevant technologies.

this is what our company requires of itself, and a good company organizational model has been established around this goal. when i first arrived in silicon valley, i observed a problem,some businesses that call themselves technology companies are not organized that way.

in the companies i just talked about, the ceo is not a technical person, and there are no technical people on the board of directors. one person on their management team is the director of engineering, and everyone else is non-technical. if a company's team is like this, it cannot be called a technology company.

i pay attention to diversity in management within my company. executives who manage different product groups of the company have different technical development and promotion paths, and a balance needs to be struck between them. we certainly don’t want all executives to be engineers. other types of talent are also important.but if engineers don’t have enough voice, then the company is not a technology company.the same is true for boards of directors.

we can smoothly switch from one platform to another and accomplish all kinds of different things because we invest in and care about the underlying technology. this is fundamental.

the product experience we build on this technology is a superstructure. we don’t think of product strategy as one specific thing, but rather how we should learn and improve our users’ experience as quickly as possible. i will personally determine the company's strategy. we also learn faster than other companies. i believe we will win.

our company is curious and values ​​learning. we're going to build better products because we're going to launch them first or early enough to get a lot of user feedback so we also know what people like better than other companies. i think that's our secret: be a technology company, build the foundation and learn to understand the topics that people care about, and then iterate as fast as possible.

14. product innovation is both a discovery and an invention. there is no shame in learning from others’ strengths.

ben gilbert: for you, is product innovation an invention or an act of discovery? do you discover a product, use the best tools and capabilities to bring it to market and get feedback, or do you envision an entirely new product from scratch and then bring it to the world?

zuckerberg:does this have to be an alternative? i think it's a combination of both. i think we need to pick some values, whether they're values ​​that you have or values ​​that you think should exist in the world, and try to build products that align with that and make those products resonate with people.

i think if you only do the latter, you won't have enough confidence to go through hard things. if you only do the former, then you may not achieve product-market fit because you’re not paying enough attention to your customers. i think both are important.

ben gilbert: when i look back, i realize that sometimes the market just happens to discover a certain format. for example, other companies discovered the "stories" format and it became popular overnight. you also have these products with stories function, which very well behaved.

▲stories have become one of the important features of meta products (source: meta)

but everything you guys are trying to do at reality labs, the $50 billion-plus that you're investing in, it's like forcing this product to exist because you seem to want to shape the world in some way. you may not need much feedback, just to launch these products.

zuckerberg:it's still a combination. we definitely have some brand new inventions, like the first version of the feed we launched in 2006.

until then, social networks were basically profiles. we believe that people actually want the latest information, and we can show them that information, and the way to do that is by ranking the information. there is so much new information, which can help people parse new information quickly, that today it is difficult for us to imagine social products without information flow.

some things we created first, some others did, and we're proud of that because we learn from it. we're not shy about learning from good things that others have discovered first, and we can also build a better version. no one company is going to invent everything, right? but i think it's hard to be a successful company if you don't invent anything.

there are certainly more smart people outside the company than inside.if you don't learn from what's happening in the market, you're missing out on opportunities, feedback from your community and customers, and understanding what they want you to do.

15. ai and ar will become the representatives of the next stage, aiming to make “amazing” products

zuckerberg:what i care more about is always the product experience and what scenarios it can enable and develop. later i had some changes and some people told methere's a difference between building good things and building awesome things.

a good product is useful and something people use every day because it adds value to the user's life. but "amazing" is different. awesomeness is inspiring, inspiring, and makes you more optimistic about the future.

so i think most of what we’ve done on social media so far is pretty “good.” we build these products. more than 3 billion people use them almost every day. it helps people stay connected, build businesses, and form communities, but not many people wake up and say social media is “amazing.”

over the next 15 years of the company, our next phase, i hope we canin addition to making "good" products, make some products that can be called "amazing".both are important, but the latter, which is a lot of what we do at reality labs, will become the representative product of our company in the next stage.

i think a lot of the ai ​​stuff that meta is doing will fall into this category. there are a lot of things in apps that would fall into this category, including new apps, but i'm not sure. i think that's mostly a good thing, and maybe it's just a reflection of my stage in life.

i like to think of myself as young, but i'm a little bit older, but i do think at this point, it's not just a meta thing. in my personal life, i value doing inspiring things with people i find inspiring.

i work with priscilla, my wife, and a bunch of amazing people on interesting scientific questions. i get to design t-shirts with the best fashion designers in the world. i made a statue of my wife and started a cattle ranch on kauai to see if i could raise the best quality cattle in the world.

16. the epidemic gave xiao zha the opportunity to reflect, and his behavior style underwent a huge change.

ben gilbert: when did these things become a priority for you? it feels like your transformation was very sudden. is there a particular moment? or have you always been like this, but we just couldn't see it before?

zuckerberg:i think there may have been some shifts when companies changed the way they operate due to the epidemic. during the pandemic, with all tech companies temporarily working remotely, it was an interesting time that gave me more time to take a step back.

i'm quite an introvert and i do need to pay attention. i get a lot of value, energy, and ideas from being around other people, but i also need time to be alone. i kind of got that opportunity during the pandemic.

it was a period of reflection that allowed me to think about these things. our country was also going through difficult times politically, and our company was at the center of that storm, which prompted me to reflect.

i think some of the small-scale things we started before, such as reality labs started in 2014 and fair (fundamental ai research) started around 2013, also inspired me.

these things have been growing since then, and when they reach a certain level, we start to think, do we need to invest more? ultimately i think research should continue and this area of ​​research will become a very important part of the company's future.

we know it will be painful to do so. we need to build ai infrastructure and expand the scale of reality labs. many investors certainly didn't like this plan, but i didn't expect it to coincide with a recession. although we have lost a lot of market value, i believe that what we are doing is important and will gradually get better.

17. changing the name is not to avoid controversy. it has been discussed within the company for a long time.

ben gilbert: the time is almost up, let’s wrap up quickly. given the breadth of your applications, it makes sense to me to rename the company. you are currently researching ai, vr/ar and other technologies. if you were to rename the company now, would you still choose the name meta?

zuckerberg:i like meta. it's a good name. we talked for a long time about finding a good short name. because obviously facebook still continues to grow in importance in the world, and i think a lot of people don't realize that, it's incredible scale.

there was a period in the past when our company had one super app called facebook and a handful of smaller apps. but now we have four apps with over 1 billion users, and if threads continues to grow, there may be five. we had been debating whether the company name should be the name of one of its apps, because we really had a strong family of apps at that point.

some people asked the question from the perspective of fleeing the facebook brand because we were politically controversial at the time. however, i believe that we are not running away from something, but moving towards a certain goal. i decided to hold off on changing the name until i came up with a name that represented the future we were building.

when we came up with the name meta, i felt it was time. at that time, we also began to increase investment (in the metaverse field), so we decided to change the name.

ben gilbert: if i were to argue this choice to you, i think meta's core competency is that you can discover products in the world. you have great ideas and work hard to make them happen. you find interesting products,

and you are not the kind of person who wants to be defined by anything. you want to control all kinds of interesting things, maximize your freedom, see where the world is going, and then have the best ship to navigate the future.

will choose a brand that doesn’t limit itself to a specific future. i'd probably look for something a little more like "i'm going to maximize my mobility."

zuckerberg:this is actually how we have always operated. we will aim at the company's vision and goals and move towards them. our company has always been committed to what we preach. if someone puts a wall in front of us,then there will be a "small puncture-like hole" in the wall, we will definitely break through these barriers.

conclusion: xiao zha advises her daughter not to want to be taylor swift, learn from others but always be yourself

ben gilbert: i have one last question. there are a lot of developers and founders here tonight, and we're probably in the most interesting technology environment since the early days of mobile, and the opportunities are plentiful. it’s been 20 years and you might need to reminisce, but what advice would you give to founders today?

zuckerberg:i think just do the things you care about. if you want to imitate our strategies, you'll have to learn them asap. but i would say there are different ways to structure things, and our way has worked for me and our team, but different companies obviously have their own ways of being successful.

one day, i took my daughter to watch a taylor swift concert, and she said, dad, i want to be like taylor swift when i grow up.

i said, you can't, that's impossible for you. she thought about it and said, ok, then when i grow up, i hope people want to be like august chan zuckerberg (the name of xiao zuckerberg’s daughter). i thought, great.learn from the successes and failures of others, but do your own thing.

david rosenthal: i love this quote, it’s a perfect ending.