打造 Meta | Andrew 'Boz' Bosworth(CTO)
Making Meta | Andrew ‘Boz’ Bosworth (CTO)
Introducing the Guest
Lenny: You were basically the 10th engineer at Facebook. I imagine there was a lot of pain and suffering that people don’t often hear about.
Andrew ‘Boz’ Bosworth (00:00:07): I didn’t sleep for more than four hours at a time. I’d wake up every four hours and check the report and see if anyone was attacking the site. They don’t tell you about that stuff in the movies.
You worked 120 hours per week., you had no hobbies.
Andrew ‘Boz’ Bosworth (00:00:17): I don’t want to take away from the romanticism of it. It’s just that most often, we hear those romantic stories from the successes. It’s a healthy thing for people to want to throw themselves into something and take that risk, but it is not glamorous at the time.
The newsfeed. That was one of your early projects at Facebook. People did not want it. They were wrong, clearly.
Andrew ‘Boz’ Bosworth (00:00:33): Now, newsfeed was an easier case than people suspect. Everyone was outraged at the same time as they immediately doubled their usage of the product.
In terms of the economic utility, the Venn diagram of Boz, of newsfeed and ads created $1 trillion of value.
Today, my guest is Andrew Bosworth, or Boz, as most people know him. Boz is the chief technology officer of Meta. He joined what was then called Facebook in early 2006 as one of the first engineers, and during his 18-year tenure at Meta, he created some of the most impactful and important products in internet history, including the Facebook newsfeed, which was the first ever algorithmically ranked content feed of any social network, and is basically what people think of as Facebook today.
He also built the original Facebook mobile ads platform, which he then ran for another four years. He also helped build and scale the Facebook messaging system, the profile, the timeline, Facebook groups, and even the internal engineering boot camp. Most recently, he served as VP of ads and business platform, where he led engineering product, research, analytics, and design. And in 2017, he created the company’s AR/VR organization, now called Reality Labs.
These days, Andrew leads Meta’s efforts in AR, VR and mixed reality, along with consumer hardware across Quest, Ray-Ban, Meta smart glasses, and more.
In our wide-ranging conversation, we touch on so many important lessons and stories, what it was really like in the early days of Facebook, why you should be asking your manager for help more often, why communication is the job. Lessons from Meta’s turnaround over the past couple of years, Boz’s thoughts on the Apple Vision Pro, a bunch of leadership and career advice, what it was like to build the very first newsfeed, and lessons from that experience, and stories of failure and stories of success, and so much more.
If you enjoy this podcast, don’t forget to subscribe and follow it in your favorite podcasting app or YouTube. It’s the best way to avoid missing future episodes, and it helps the podcast tremendously.
With that, I bring you Andrew Bosworth, a.k.a. Boz, after a short word from our sponsors.
Boz, thank you so much for being here. Welcome to the podcast.
Andrew ‘Boz’ Bosworth (00:04:55): Thanks for having me. I’ve been a long-time fan of your program and all the things that you’ve been putting out, so I’m glad to finally get a chance to join.
Same. I’m really excited to have you here. I have at least a billion questions I want to ask you. But I want to start with a few fun facts that I’ve found about you, and what if I go through them and then just pick one that stands out, and then tell the story behind it? How does that sound?
Andrew ‘Boz’ Bosworth (00:05:17): All right, sounds good.
Okay. You went to 14 proms.
Andrew ‘Boz’ Bosworth (00:05:22): Yeah. It’s true.
Okay, I’m going to keep going. Okay.
Andrew ‘Boz’ Bosworth (00:05:25): Wow, that’s a strong opener.
That might be the one. You were a national Taekwondo champion in college. You were Mark Zuckerberg’s TA in college in a class on AI, which isn’t actually how you landed at Facebook, from my understanding. Harvard was recruiting you to play football for them. You were very active in the 4-H Club, and you raised animals and showed them at county fairs when you were growing up. You once shared a stage with David Copperfield.
Andrew ‘Boz’ Bosworth (00:05:52): That’s true.
MC Hammer once told you that your outfit was stylish. And president George W. Bush complimented you on your shoes and the shine of your head.
Andrew ‘Boz’ Bosworth (00:06:03): Yeah, these are all true. I want to say, wow. First of all, I got to make sure that people understand, I was a national collegiate champion, in as a green belt, which that’s like being the JV champion. Just so everyone’s clear on what that is.
W’s a W.
Andrew ‘Boz’ Bosworth (00:06:16): Heavyweight sparring. I’ll tell… The prom story is a funny one. It’s related to the 4-H story. I was a big-time 4-H’er, National 4-H Hall of Fame, did all this stuff there.
As a comeup to that, 4-H is a wonderful program, youth program, it’s coed program, and I was all over the state, all over the country, doing leadership events and doing these conferences, and doing a lot of public speaking.
And almost every 4-H event has a dance. People don’t know that. At the end of a conference, at the end of the, literally camp, you’d go camping for a week, at the end there’s a dance.
And so, as a consequence, the most important thing, if wanted to go to a lot of proms, I was a good dancer. And it turns out, the high-level bit, at least in the 1990s, for girls selecting who they might want to go to prom with was, will he and can he dance? And the answer with me was yes. And combine that with the fact that I knew a bunch of girls who went to different schools, that’s the recipe for success right there, if that’s the goal that somebody has. Two my junior year and 12 my senior year. I once went to three in one weekend, a Friday, a Saturday, and a Sunday night.
The Opening Anecdote
Lenny: Another fun fact about you is you were basically the 10th engineer at Facebook, initially, way before it was a clear success story. I imagine there was a lot of pain and suffering and struggle that people don’t often hear about those early days. They see a movie like The Social Network. It looks like, “Oh, that was so much fun. I’ve got to start a company. It’s going to become a trillion-dollar success story.” I’m curious just what those early days were like. Are there memories that stand out to you?
Andrew ‘Boz’ Bosworth (00:07:47): Yeah, there’s a bit of a joke in the 10th thing, which is me and five other guys all joined at the same time, and there was nine engineers before us. We joined the same day, so we’re all the 10th engineer. So, somewhere between 10 and 16, depending on how you want to do the numeration on that.
I’ve written about this in my blog, and I tell this story a lot, which is it was fun, and there was tremendous camaraderie and memories that were formed, but they were formed in a kind of forge of really intense times.
At that time, almost all of us lived within one mile of the office. We ate most of our meals together because we were working, not to say we weren’t also friends, but because we were working, it’s like, oh cool, you just roll into a meal and roll back into work.
And there’s little things that you don’t appreciate, which is like there was nobody to help you. There was no expert. And so, it wasn’t like, “Hey, I’m struggling with this one tricky problem. Who should I talk to?” It’s like, nobody. You should talk to yourself and figure this out. Or it’s like, “Oh, man. My servers are out of capacity.” It’s like, “Cool. You should go to Fry’s Electronics, you should buy a bunch of components, you should build a new server, and then you should run it. Maybe drive into the colo, rack it, and then get back and run it.”
People really undervalue the fact that when you go to work, even a moderate mid-size corporation today, especially with the tremendous growth of startups supporting startups, things like payroll and finance and IT and HR, these things are professionally handled in many cases.
That was just not the case in the early 2000s. It was just you, and your personal car, and whatever you wanted to do with your time.
So, I don’t want to take away from the romanticism of it, it’s just that it’s most often we hear those romantic stories from the successes. We so rarely hear somebody who went through really sacrificing a lot of my 20s from any kind of social or outgoing, fun environment.
It paid off for me, so no one feels bad for me, nor should they. But there are other people who do the exact same thing, maybe they worked harder, maybe they were smarter, maybe they did better, and it didn’t play out for them. And it’s a big sacrifice.
And so, I love the enthusiasm for startups, I love startup culture. I think it’s a healthy thing for people to want to throw themselves into something and take that risk. But it is not glamorous at the time. In retrospect, it’s like, “Oh, it can be…” We have a little halo around it. But at the time, it doesn’t feel glamorous.
Yeah. In this post you mentioned, you said that you worked 120 hours per week.
Andrew ‘Boz’ Bosworth (00:10:07): Yeah.
You had no hobbies, and you gained a lot of weight.
Andrew ‘Boz’ Bosworth (00:10:15): Yeah. We drank a lot to make up for it, so I gained a lot of… And you weren’t eating healthy. It was crazy.
I think I’ve told this before. There was a time where one of the first things I built was an anti-spam, kind of anti-scraping defense mechanism. But we didn’t have any ops support. There was no 24/7 online support.
So, I built this tool. I had to wake up every four hours. For about two years, I didn’t sleep for more than four hours at a time. I had to wake up every four hours and check the report, and to see if anyone was attacking the site. And if they were, I was up, and I had to go battle back. And if they weren’t, cool, I’d go back to sleep. But that’s the thing, they don’t tell you about that stuff in the movies.
That’s almost worse than having a kid, a newborn.
Andrew ‘Boz’ Bosworth (00:10:59): And nobody asked me to do it. Nobody even asked me to do the anti-spam, anti-scraping stuff. I just thought it was a problem, and I went and did that, and that was the solution I come with. If I was a better engineer, maybe I’d have solved a better problem.
So, maybe just to close out that thread, when you talk to founders, what advice do you give them along these lines?
Andrew ‘Boz’ Bosworth (00:11:18): I want to be cautious about this, because what I tell founders… The first thing I tell founders is that I’ve never been a founder, and I want to recognize the difference. I joined January 9th, 2006. That’s almost two years after Mark started the company. I wasn’t involved with fundraising, I didn’t have to do any of that side of the things, and I didn’t even have to deal with the board or the business side of things. I really was lucky, in a way, to have joined when I did.
But the first thing I tell founders is like, “You should take my advice with a grain of salt. I have not actually been in your shoes.”
If I can compliment you, really, one of the things I like about your program is, there’s a whole system of professionals in our industry. And when I grew up in technology in the valley, you always heard about the ACM, right? The Association of Computing Machinery. You heard of these legendary professional organizations that supported people in our fields.
And by the nature of the rapid pace of change in the technology, and the nature of the engagement of those institutions, even academics, even academia broadly, kind of are out of touch. The tools that you got from those places weren’t useful in our field.
So, I do think the mentorship that we give each other has been a critical and sustaining resource. There is, today, now, resources like your podcasts and your newsletter that are actually really designed to help people who are professionals in our industry in a way that is almost kind of missing for 15, 20 years, and I love to see that. Because if you’re an up-and-coming [inaudible 00:12:45], literally you used to have to know somebody and ask them a question.
And so, a lot of times what I’m helping founders with, I can help them with the strategy, I can help them think through the technology choices, I can help through business, I can think through the management, the organization structure. But I also try to be very clear, there’s a bunch of stuff that I just was never exposed to.
So, even as we just talk about how tough it was for the average engineer joining Facebook in 2006, man, it was even tougher for Mark Zuckerberg in 2004, probably. And that’s a story that’s been told, I’m sure, but still. So, I think these are both… It’s almost all scale and variant. No matter how far you dial back, the challenges are interesting and are worth talking about.
The Prom Dance Master
Lenny: One of maybe your most popular posts is this quote that you share about the advice you often give. What you say is, “The advice I find I have to give more frequently than any other in my career, as a manager, a board member, an advisor and a friend, is for people to more directly leverage their leaders.” Can you talk about that and what that means, and what that looks like?
Andrew ‘Boz’ Bosworth (00:13:40): It’s such a normal and natural healthy thing. And by the way, we do it in our personal relationships. Like I said in the post, we want to do it ourselves. We want to do it ourselves to prove to everyone that we can do it ourselves.
And we think in our heads, “If I ask for help, haven’t I already given up that goal? Haven’t I just admitted defeat on one of my top level goals, which is to demonstrate that I can do it myself?”
But what so often we forget is, more often than not, your job is not to do it yourself. Your job is to get it done, is to have the thing done, done well, done right, done competently. And a lot of times, the tools that you need to do that live with your manager, with your partner, with your advisor, with your mentor. That’s where they live.
So, how many times as a manager have I gone through, and somebody’s told them, “Here’s the job.” They’re like, “I got it.” They go off, they come back, it’s done, it’s wrong. And I’m not blaming them for it being wrong. They didn’t check in with me. They misunderstood. We miscommunicated. I’ll take the L on that. That’s no problem.
But here we are, six months later, it’s not done right because they misunderstood the brief. We miscommunicated with the brief.
Or they come back, and it took a year. And I’m like, “Why did this take a year instead of six months?” And they’re like, “Oh, man. I had all these things I had to deal with.” Where, if they had emailed me, I could have bulldozed that stuff. I could have cleared the path. I could have said, ” Oh, no, no, no. Don’t worry about that. This is the thing.” And then we’d have been done six months sooner, and they would’ve been less frustrated. And so, light touches.
Now, I do think as a manager we also have a job to say, “Hey, that’s kind of the work, so you got to kind of go figure that out.” And one of my things I always tell my managers, one of the most powerful things we do is refuse to rule. Someone will bring me a thing. A lot of times we feel obligated to weigh in and help. I’ll be like, “Nope, but look, I think you’ve got it. I think the challenges you’re facing are the right challenges. I think you’re approaching it in the right way. Just do your best there.” And that’s what it is. And so, there is a responsibility as well for those of us who are leaders, mentors, advisors, board members to do that.
But, by the way, we do this… Personal relationships. You’re with your partner, and you’re trying to do something the right way, but you’re not talking to them about it. You’re just taking a huge risk there, and for very little reward. They’re not going to be mad if you ask them, like, “Hey, is this how you wanted it done? I don’t know.”
And so, I do think it’s kind of funny how much we build these castles in our mind, these little silos that keep us from engaging the structures that are built around us, that are designed to help us succeed. I saw this great quote, actually just yesterday. I saw Patrick Stewart, who is one of my favorite actors of all time, and whose characters I love, and he talked about people going on casting calls. And this is a brutal thing for actors, right? You’re going on 30, 40 things, you’re getting rejected. It’s tough. Everyone’s kind of heard about this. And he said, “No one wants you to succeed more than the person you are auditioning for, because they want you to be awesome. Because as soon as you are awesome, they’re done. They want you to be amazing.”
That’s like your manager. Nobody wants you to be more awesome than your manager does, because when you’re amazing, your manager, his life gets easier, her life gets easier. So, I just think that’s the mentality we get into is like, “No, no, no. They’re testing me.” They’re not. They are rooting for you. I promise you that.
I love that advice. I imagine the reason people don’t do this, as you said, is they don’t want their managers to think they don’t know what they’re doing, or they can’t solve it. Do you have any advice and guidance for when it makes sense to go reach out and ask?
Andrew ‘Boz’ Bosworth (00:17:06): One of the things I think, for people who are timid about this especially, is I think you can put a framing around it that’s really easy for your manager to engage with. You can say, “Hey, I’m making progress on this. This is what I’m blocked on, this is the current program.” And I’ll even say, “Hey, if this all looks good to you, no response required. If there is something here that you want me to do better, different, that you think you could help with, let me know.”
I love a typed, five-, 10-sentence email, “No response required. Here’s where things are.” Even if everything is going really well, I’m like, cool, this person understands the urgency, they understand the assignment, and they’re giving me a little heartbeat, a little ping back.
And then also, if two weeks later, let’s say the blocking issue is bad, then you say, “Hey, I am sorry, but I do actually need your help now. I’m actually blocked on this thing.” I have the context. I have a mental model of you toiling away on the right thing, on the thing I asked you to do over there. Even then, when you’re blocked, you can make my life super easy. Like, “Hey, what I’d love for you to do, if you could send an email to this person, here’s a draft with this thought, that would help.”
Or it’s like, here are specific questions framed up. “I think this is what you want. Is this right? Yes or no? If no, okay, we’ll come back and we’ll spend more time. If yes, we’re all good.” So now, not only am I up to speed, I have a mental model, I’m engaged. Also, you’ve made it super cheap for me to help you. And people are always surprised. People who work for me are always surprised when I tell them how big a part of my job is doing these little types of things.
It’s a little spinning plates at my scale. I’ve got 10,000 or 15,000 employees, depending on how you want to count different things. And so, you’re just like, every now and then I got to get a whole new plate, a whole new rod, and just really put the effort into it. But for the most part, I’m just trying to touch everything and keep the momentum going. And so, if something falls, and somebody didn’t tell me that “Hey, we’re losing rotational velocity here. We’re losing momentum.” Oh, I’m bummed. I’m like, “Ah, now that plate fell. I got to start a whole new thing over here now.”
So, I think people underestimate. They think of my job differently than my job actually… My job is actually tons of little touches.
So, I think a key takeaway here is, one, index more towards asking your manager and leaders for help. And I love this way of framing it of, it doesn’t always need to be like, “Here’s what I need from you.” It’s, “Here’s what’s happening. Here’s things that might be blocking me. Here’s questions I have. Here’s things that are going on.”
This is actually similar to something I found really powerful that I’ll share real quickly, this idea of a state of… I call it the State of Lenny email. And I sent this email to my manager every week. The State of Lenny. It’s kind of like State of the Union. And it’s, “Here’s my current priorities, here’s what’s on my mind broadly, and then here’s blockers that I need your help with.
Andrew ‘Boz’ Bosworth (00:20:02): We actually used to have a format for that we called HPMs. Highlight, people, me. And every manager at Facebook from like 2008 to like 2014 would send to their manager, or even their leadership group. I mean, at one point when I was running what we called comm apps, I just sent it to Zuck and the whole leadership group.
It’s like, what’s the highlights include? And it highlights [inaudible 00:20:27] flow. What’s the big ticket things you need to know? Where are people? Like is somebody in trouble? Is somebody at risk? Is somebody doing really amazing work that needs a shout-out? And then me, how are you personally doing? HPMs, we called them.
Actually, it’s funny. I hadn’t thought about that in a long time. But yeah, no, I think this kind of thing can work. And look, every manager is different, so even at the Meta level, by the way, is another success. I think what people do is they want to treat every manager the same, and that’s not going to work because every manager is different.
But every manager, you can ask, “How do you like to get updates?” You can ask them when you first start working with them. Like, “Hey, what’s your cadence? How do you like to stay informed?” And so, for me, I do regular one-on-ones. As the org’s has gotten bigger, those have gotten more distant, so people have replaced those with more written things.
But no manager will get pissed at you if you’re like, “How do you like to get information about me?” That’s a totally reasonable thing to ask.
The Reality of Early Startups
Lenny: I love the specific idea you shared of just drafting the email to say the other team leader, “Here’s what I need you to tell them. That would really unblock this thing.” And that’s such a cool idea.
Andrew ‘Boz’ Bosworth (00:21:27): By the way, I always put my own… I don’t take that copy paste it. I’m always looking at that and be like, “Okay, I understand.” A lot of it is actually not about what you want the other person to hear, it’s about the voice, the tone.
It includes a lot of history. I don’t know. Have you been going back and forth with them for 12 rounds, and this is going to feel to them like I’m really coming over the top? Or is this like, “Hey, first time you’re hearing about this, my bad. Here’s what we’re doing. Need your help.”
So, a lot of that isn’t even about making my life easy because I want to copy paste. A lot of it is, actually, there’s a rich set of information in how you tone and how you draft that note that’s going to help me land it correctly and not feel like I’m just out of band, heavy coming in.
This touches on something that I often hear is very core to the way Meta works, which is transparency. Anyone can ask Zuck questions at the Q&A’s. People are encouraged to post constantly internally of what they’re thinking, what they’re working on. All the data is shared publicly. Which often leads to leaks, which I hear you hate, and that is a pet peeve of yours.
Andrew ‘Boz’ Bosworth (00:22:32): Just feels like a violation of the team trust. Just feels like… I grew up with playing sports, right? I was football, soccer, track. And you just can’t imagine one of your guys calling out the play to the other team. Can you imagine what you would do in that case? “You’re off the team. I’m sorry, you can’t be here.” Anyways, sorry. Carry on.
Yeah, and there’s so many more people, it’s hard to find. Who is this? So, with this downside as an example, and I imagine there’s other downsides also takes a lot of work, and it puts people on the spot a lot of times, what have you seen as benefits and why is that such a big part of Meta’s way of working?
Andrew ‘Boz’ Bosworth (00:23:05): Yeah. This kind of comes back to, I think, the principle that really good, talented people, you want to leverage them fully. You really want to make sure that they are fully leveraged. And so, anytime they have the wrong information, or they don’t have the information, you’ve now blocked one of the economically most valuable things that your company possesses, which is this person’s time, attention, talent. Not only that, you’ve also made them more frustrated, and now they are more likely to leave.
If the lifeblood of any company are the people inside of it who collectively commit to some kind of a goal and a mission and work together, then you want to maximize that potential. And creating this really open information ecosystem is one of the ways that we do that. So often, great, phenomenal work that has happened at our company has not come from this one top-down mandate, but has come from people understanding not just what we’re trying to accomplish top down, but also having way more information at their disposal to be able to act on it.
People talk about top-down or bottoms-up culture, and it’s a bit of a myth, in my opinion. If you’ve ever met Mark Zuckerberg, it’s not a bottoms-up thing. The ideas that we’re pursuing are Mark Zuckerberg’s ideas, first and foremost. That’s not to say that he’s not open to new ones, and of course he is, and that’s a form of bottom-up, people can bring ideas to him and he internalizes them and acts them or not. But when he brings things top-down, he’s not micromanaging, he’s in the details. I’ll be careful on that.
But he does create the space for you to bring back three or four versions of the thing that he’s talking about, and then he shapes it from there. And you can’t do that if there’s… If you don’t have degrees of freedom, sure, but also if you don’t have the information. Otherwise, if you don’t have the information available, what we’re trying to accomplish, why we’re trying to do it, what the infrastructure is like, what the availability is like, what the performance is going to be like, well, you just are stuck. You’re just going to have to follow the script. That’s very boring for high-talent, high-creativity, high-engaged people.
Now, it does come at a tremendous price. You have to get really good at managing information on the incoming. Most people at most companies consume all the information that’s given to them, but the information itself is carefully managed. They’re getting all the information they’re intended to get.
We’ve turned that on its head, and it sounds great, but it’s not free. Even somebody senior coming from outside of the company to the company, one of the things I have to coach them on is, how do you find signal amongst all the noise?
You have to have a system for managing your information. You have to have a system for triaging the incoming, getting rid of a bunch of stuff that is on the wrong channel, or it doesn’t matter to you, figuring out what’s the… groups that you want to be a part of, but you consume only when you choose to and-
Andrew ‘Boz’ Bosworth (00:26:00): … groups that you want to be a part of but you consume only when you choose to. And where are the things where you’re getting push notice? Like that’s the real time thing, and that takes some time.
This point you made about bottom up versus top down is really interesting, because, when I think of Meta, I think it’s a very bottom up culture. I hear everyone comes up with their ideas, runs experiments, is very encouraged to just try stuff, and it’s really interesting to hear that. And it makes a lot of sense that most of the big ideas actually do come from the top.
Andrew ‘Boz’ Bosworth (00:26:26): It’s of these mythology things. I don’t think it’s the wrong… As a contract, it’s more bottoms up than many other companies, because you do have these degrees of freedom within the construct. But make no mistake, like Mark or me or Chris Cox or Javi, they have very strong opinions about what we should be doing as a company, and your bottoms up-ness works within that framework. But it is true that you can ask Mark any question, and he’s going to answer it. Same with me, same with Chris, same with Javi and also that we certainly take inspiration from the discussion that we have with employees. So I don’t know, it’s just not as black and white as people tend to paint it.
Advice for Founders
Lenny: I think one of the biggest lessons here is making it at least feel like you have a lot of say, even though a company is very, “Here’s the big strategic [inaudible 00:27:10],” you’re very good at making people feel like they can have an impact and a say.
Andrew ‘Boz’ Bosworth (00:27:14): And can I tell you, the most important thing is just giving people clear guidelines so they know where they… Like, “Where do they have space and where do they have no space?” One of the things that we go in these reviews with Mark or my team with me I’m sure, and I’m like, “For this part of the UI, it is going to… Like, I will draw it for you. It’s going to be like this.” And this other part. I’m like, “Cool, that’s important, too. I don’t have a clear vision of it. Why don’t you do it?” So there’s just really clear guardrails of like, “Okay, where are we just on assignment? And where do we have more flexibility?”
Is there an example of that that comes to mind where you were just very in the details and drawing the screen?
Andrew ‘Boz’ Bosworth (00:27:51): Over the course of time, there’s been quite a few examples. I think early on, when we were working on News Feed, mark was absolutely whiteboarding every single pixel that the team had to put on the front end. On the back end, he was like, “Make it rank, have some ranking.” So I felt like I was lucky there. I was just like, “Cool, I’m off to the races on some ranking stuff.” And Chris Cox and all these other guys are having to really pixel-match these things. But it’s not always that way by the way. So now fast-forward and we’re talking about ranking, it’s not like Mark is always hands-off on that. When we got into modernizing our ranking systems, which we’ve done over the last five years, Mark was heavily involved and like, ” Hey, what’s the mix shift and how are you weighing different things?”
And so it can go both ways. For me personally, I’ve gotten really involved in some relatively esoteric things. I was really adamant, for example, that hand tracking and mixed reality make it into the headset. Let’s just say that there weren’t any supporters in the team. Obviously we had a hand tracking team, which is phenomenal, mixed reality team, but there was a lot of people who were like… They did not feel those features were going to be critical for this to become a mainstream device. I always believed that they were for ease of use and for… So I just really forced the issue and didn’t give anyone any room and held really high standards for the performance benchmarks we were going to hit on the hand tracking. And teams told me it was impossible. And it wasn’t. It did great.
This touches on something that comes up a lot on this podcast. And there’s this debate between how in-the-weeds founders and execs should be, whether they delegate, empower, versus, “No, we’re just going to do it this way. I’m going to be very involved in every mock.” And there’s always this up and down that happens where it’s like, “Okay, cool, we’re going to let people run and do their thing,” and then things start to not work as well often. And then, “We’re going to take back control.” You have just a perspective on when it makes sense to go deep, how founders execs should think about that?
Andrew ‘Boz’ Bosworth (00:29:41): Such a useless answer for founders, it depends on the weeds. There are some weeds that really matter and there’s some weeds that really just don’t, and I should say, that doesn’t mean they don’t matter at all. You have to do them, but they aren’t the hinge upon which success or failure will happen. Yeah, there’s people I really respect, Brian Chesky’s been on and said, “Look, the Airbnb is going to work only on the things that I can work on. It’s that’s the extent of what it’s going to do.” And that’s a super extreme form of it. I have a lot of respect for him and how they’re working things. I think that if you have great super talented people that you can trust who can own bigger pieces, that’s one option. If there are ways to structure it so that you can check in effectively and make sure that it’s on track, that’s another way to structure it.
And there probably is still work happening at Airbnb, that has to happen, finance and accounting and HR, that Brian isn’t personally managing. So there are clearly non-technical areas that we do. Or legal. There are areas that we do trust that this is happening at. And so I think a lot of founders regret delegating too much, from my conversations. And I totally get that. Or they delegated something critical that really turned out to be the most important thing. For me, the judgment is like, “How do you, most important, determine what is what matters the most?” And so Mark, we joke inside of Meta, to this day, we call it the eye of Sauron. When Mark has determined that the thing that you’re working on is the most important thing, there is no detail too small for him to notice.
He will be in a review, and in the same review. We’ll be like, “Strategically, I think we’re off course. And also, this one pixel is definitely wrong. You have to fix that.” That’s a big range. Frankly, I’ll be a little bit self-congratulatory, I pride myself on being able to do the same. And I think people who work with me often comment that the style of leadership that we have, and I think Chris Cox is the same, is where it’s like we will go where hide a low on the things that matter a ton. And there’s a bunch of other things that certainly matter. We’re glad we’re doing them, but either they have pretty clear roadmaps, pretty clear examples in the industry, or it’s like that’s a feature that you have to have but isn’t going to determine success or failure. So getting it into rough shape and then iterating on it is fine. And so I think it really does depend on the weeds, how deep you want to get.
It’s so funny, I use exactly that same metaphor, the eye of Sauron, when I talk about working on things that Airbnb that matter a lot to Brian. And my advice to people is, “You don’t want to be in that eye of Sauron for too long in your career, because you’re just going to burn out if you’re working on the most important thing all the time, but want to be close. You don’t want to be in the Shire, but you want to be around the [inaudible 00:32:26].
Andrew ‘Boz’ Bosworth (00:32:26): That’s right. They’re both… So I worked for years in ads. From 2012 to 2017, I ran ads in a business platform, this big ads group, and it was an area where certainly it was very important, but Mark had so many other things going on with the transition to mobile, he did kind of delegate it to me. And it was awesome, and it was so cool to have that kind of trust from him. And also, you’re constantly terrified, because like, “Mark does not know. What if this is all…” My leads would be worried, because they just hadn’t had a review with Mark in a while. And it’s like, yeah, you suffer in the intensity of the gaze of Sauron. You also suffer in the shadow of its absence. There’s no perfect place to be.
That’s hilarious. I’m trying to think of the part of Middle Earth that is a metaphor for that.
Andrew ‘Boz’ Bosworth (00:33:14): Yeah.
Top Advice: Leverage Your Boss
Lenny: Okay, so you talked about the News Feed, which was one of your very earliest projects at Facebook. Here’s a couple fun facts I know about the News Feed. One is that it was the very first algorithmic News Feed of its kind of any social network and maybe of any sort of product like this. And two, it was the very first AI code that was written at Facebook to rank the actual News Feed. So there were a lot of firsts, and clearly this became a huge deal. The News Feed is essentially what people think of when they think of Facebook now, but it was super controversial when it came out. People were very against this. They did not want to be sharing this much information with people, or so they thought, and then they realized eventually, “Oh, this is actually exactly what I want.” What did you learn from going through that experience of building something that people initially reject and then later realize that they actually do want this and this is exactly what they’re waiting for?
Andrew ‘Boz’ Bosworth (00:34:08): This is a story that you tell a lot actually through your interviews, which is just like, “You have to have conviction in what you’re building.” You’re choosing your customers as much as your customers are choosing you, is one way I think about it sometimes, and one mistake that you do see sometimes startups make is they get an early cohort of users whose needs actually take them kind of orthogonal to a larger market. And so they become kind of held hostage by their earliest customers now. So time and time again, we’ve had a vision for what we thought this should look like, and it wasn’t the thing we were delivering right now. And so people who were using the thing we currently had were not sure that that change was what they wanted. But we had a confidence that over time they would, and we’re not always right, but in these cases we were right. Now News Feed was an easier case than people suspect, because everyone was outraged at the same time as they immediately doubled their usage of the product.
So we had a few advantages there, which was, it was literally like everyone was like, “I hate this so much.” And they would refresh, refresh, refresh. And so we were like, Okay, wait, there’s cognitive dissonance here between what the stated preferences and what the revealed preferences are in the economic sense. So News Feed was a little easier than people suspect, to us, to stick with. But people sometimes misunderstand that. They think, “Oh, the lesson is don’t listen to your customers. Not at all.” And we certainly care a tremendous amount. And even with News Feed, we did actually screw some things up. I kind of always make this joke that it’s almost like… You know when you’re at the party and music’s loud, you’re talking to somebody and the music cuts out right when you’re saying something at a super high volume, and so everyone in the party hears the last thing you said?
Now you were saying it in a public place, so it wasn’t like it was a private comment, but you also didn’t mean to broadcast it at that volume. We kind of did that to the entire user base. We took what had been wall posts, which sure, anybody could have gone to that profile and seen and then put it kind of on blast, on Main, as the kids say these days, “Put it on Main,” and someone’s like, “Ah.” So we did that. I don’t want to say it like… We did screw things up. It wasn’t like, “Oh, this is a clause execution.” So another thing to know is like, “When did you screw something small up, and when did you do something big up?” When is the thing itself wrong, versus when were the details wrong? That is an art. That is a real art, and you don’t always have user data to determine it like we did.
And so a lot of that is, “Do you have a clear vision and intuition for what you expected to happen?” And then what happened instead and can you diagnose the delta there? So in the News Feed case, we made a bunch of little mistakes. The thing itself was right, and I really am quite proud of the work we did there. Me and Chris Cox at the most core probably in the engineering side, which you saw as the PM, there was no ranked feeds before that. We did have some AI that I built before for the anti-spam, anti… things, but it was pretty rudimentary, but it was probably the first consumer AI that was in a website of that kind around content. And we built the most efficient monetizing surface in history outside of Search, I think. [inaudible 00:37:04] for those who are curious, I don’t use monetization because I think money’s the most important thing. I do think it suggests the economic power you’ve created, which I do think correlates very strongly with human utility. Although obviously I respect that some people may disagree.
In terms of the economic utility, the Venn diagram of Boz, of News Feed and ads, create $1 trillion dollars of value.
Andrew ‘Boz’ Bosworth (00:37:24): It’s not-
Well done.
Andrew ‘Boz’ Bosworth (00:37:25): It’s not nothing, not nothing.
[inaudible 00:37:27].
Andrew ‘Boz’ Bosworth (00:37:27): We’re proud of that work.
You have this quote in one of your posts about the News Feed where you said, “It consumed me more fully than anything in life had ever consumed me. It opened up to me the truth that when you’re passionate about something, you do better work, you do smarter work and you’re, in order of magnitude, more productive.”
Andrew ‘Boz’ Bosworth (00:37:43): There’s no substitute for it. One thing I’ve learned about myself since that post actually is just the degree to which I am somebody who is inclined to be passionate about things. And it’s a gift that I’m lucky to have, and I understand that’s not every person. And so actually the ads thing is a good example. When Mark told me to go work on ads, I was like, ” No, I don’t want to. I don’t think I have a passion for that.” I had this idea of myself. I had a very strong identity of myself as this AI infrastructure product guy, and I was working in this space and nope, I was wrong. I just am a guy who gets excited about things. Once I get into ads, it’s like, “Oh, this is fascinating. It’s a three-sided marketplace, and there’s all these different…” It felt like I was playing chess times in terms of the moves with the other players in the industry, and I was super pumped about that.
And then when he wanted me to work on hardware, I was like, “No, I’m not. I’m a software guy. I’m a software guy, Mark.” And now, I love this work. That’s such a fascinating space that I’ve learned so much. So I do think that’s right. I do think what I find something I’m passionate about that’s good. What I have learned since then is to give myself a space to understand if I can get passionate about it. Now, there are parts of jobs that I’ve had before where I just never found the passion, and after six months I just have to move on. Literally it’s like I’ll either quit, get fired, I’m doing bad work, I don’t care about the work. And so I do have a self-awareness. It’s not that I can get passionate about anything, but I do have a pretty broad palette, it turns out.
I think that’s a really interesting career lesson of, “Don’t assume you won’t be excited about something that may come up.” Is there anything there that you’d share with folks of just like, “Explore it, give it six months, see if you can get excited about it?”
Andrew ‘Boz’ Bosworth (00:39:14): Absolutely. So I have a very unusual career arc in some ways, which is I really almost changed jobs every six months for a long time. I was working on this integrity stuff with News Feed in the background. Then I was working on News Feed for about a year. Then I worked on site speed and infrastructure and detecting SAVs and issues. And then I worked on Bootcamp, and then I worked on messaging and groups. I had this really funny thing. I was kind of joking it was like, for those who are old enough to remember Karate Kid, I felt like I was painting a lot of fences, waxing a lot of cars, and at the end I knew karate. I was like at the end, at the end I had the payoff, because I’d gone through and I’d met a lot of people and I’d worked in these different areas and I understood different dynamics. Well, other people who joined in my cohort were getting promoted, but they were in a single track. They just stayed in one place and they got promoted, whereas I kept moving around. And it probably at some point early in my career felt like I was moving more slowly relative to my peers. And then when I finally turned the corner, really with the ads appointment, which I did for five years, I went vertical. I just like my career went vertical. And since then I’ve kind of been on that trajectory. And so advice I most often give people, for me at least, the lesson that I take from this, is I just was willing to learn aggressively. I would move because I wasn’t learning enough. I was bored, and so I wasn’t learning enough new stuff. And what was cool about finally getting to the ads job, and likewise in the job I’m in now, is those jobs, I learned a ton for five years. I’d never stopped learning in those jobs. You will occasionally find those jobs where they’re super deep and you can just keep learning.
Meanwhile, a lot of my friends whose careers were on a better trajectory [inaudible 00:40:55] then earlier, they literally got bored of what they were doing, but they didn’t have any place to jump to. There wasn’t some other… They’d become domain specialists in a domain that they’d kind of exhausted for themselves. And maybe they’d even stuck around longer than they wanted to because it was comfortable or because the company wanted them to. And it ended up kind of being a hindrance to them in the middle of their career. And so for me, it’s like, “Jump into new things, give it six months. If it’s not the thing, no problem. You just built a ton of new skills that’s going to come in handy, I promise you that. Keep going.” And likewise, when you do make that jump early career, optimize for learning, optimize for… Think about a compound interest. It’s like the first 10 years of compound interest don’t look that impressive. It’s like after 10 years it starts to look good.
How to Ask Managers for Help
Lenny: I love that advice. It’s similar advice I always give of, “Variety of experience often ends up being the most valuable thing you build over time,” just trying to bunch of stuff, doing some internal tools, maybe working on customer support, I don’t know, trust and safety, user-facing products, infrastructure. I’m thinking from a PM’s perspective, maybe an engineers, and other functions. One question along those lines. So we talked about the eye of Sauron and working on the most important thing at the company. Do you have any advice on how much of your career you should be working in that center?
Andrew ‘Boz’ Bosworth (00:42:12): Yeah, listen, [inaudible 00:42:12] being equal, I think there’s two really good places to be. I think one is carrying a lot of water in areas that the company’s not paying attention to but you are important. And it needs to be a lot. You got to own that stuff and really move mountains over there, because I promise you, as an executive, when there’s a huge dam holding up the floodwaters, you respect the heck out of the person who is holding that dam up. You’re like, “You keep doing that, Atlas, that is good work over there.” The second-best place to be, or maybe equally, is on the most important thing. And on the most important thing, that’s where you get to the advice that Eric Schmidt gave Sheryl Sandberg, which is like, “Hey, it’s a rocket ship. Get on. Don’t ask what seat I’m in. Just get on.”
If it’s the most important thing, you’re going to get a smaller piece. Everyone wants to be there. Get the piece. If it’s the most important thing, get the piece that you can crush, kill, do a great job at and grow from, because you’re going to get a ton of visibility, you’re going to get a ton of experience. You’re going to see what it looks like in the fire, in the fire, and that is invaluable. You will use that everywhere. And so I say that. That’s at project selection time, but now I’ll be cautious. Understand, projects that start in the fire, hopefully, are forged in some manner of metal that cools and is no longer in the fire, like God willing. And likewise, things like dams that are holding up floodwaters have a tendency to crack or break or floods overcome [inaudible 00:43:42]. So I think you do want to be at selection time in one of those two places, but then you also… You’re going to stick with the ride.
And again, to my point, if you’re not engaged, if you aren’t doing great work, if you don’t love it, then move on. If you’ve exhausted it, you used to love it, but you don’t anymore, move on. If you still love it and you’re engaged, great. That’s cool. That’s a great thing. You deserve to go from the forge to the dam and back over time. You don’t have to always just keep jumping onto the latest fire. I tried to do that once, after the ad [inaudible 00:44:14]. Actually, so I spent six months and we built the first mobile ad product in 2012 and kind of saved the IPO, which had gotten pretty grim [inaudible 00:44:23]-
Yeah, I remember that.
Andrew ‘Boz’ Bosworth (00:44:24): … that point. And I told Mark, I was like, “This is so fun. Maybe you can just keep doing this, just putting me on the biggest fire every six months.” And he turns to me, he said, “Buzz, that’s not a real job.”
He’s like, “I need you to stay here and usher this forward,” which I did for the next four and a half years. And it was amazing. It was amazing. And again, I do give him such… It’s funny, I’m going to get a hard time in this. One of our biggest critics as well as being one of his biggest fans. I have both those jobs, but today we’re talking about stuff that I think Mark really demonstrates really well. And he did a great job of pushing me in my career to different places where I didn’t think I could succeed. And he saw the opportunity that made it happen.
What have you learned about that, giving Mark negative criticism, anything that he accepts? What do you learned about that?
Andrew ‘Boz’ Bosworth (00:45:08): Mark’s voracious for all information and all points of view. One of the things that’s pretty interesting, I talked earlier about how much, as a founder, I think especially, you have to have tremendous conviction. You just have to have a tremendous degree of confidence. And I think Mark is somebody who is maybe the strongest willpower of a person I’ve ever met, just in a pure willpower sense. One thing that’s interesting about Mark is you’ll give him feedback, he listens. He’s a kind person to work for, so you’ll give him feedback, and he’ll listen. Truly. He’ll most often tell you that you’re wrong, why you’re wrong. That’s just like most often. And what will happen is… It’s uncanny. It’s like over the course of the next week or two, you’ll just see shifts. And I don’t think he’s doing it…
I’ve always joked that the information gets to him. So much information every day gets him. And then at night, he re-compiles the whole world with all that information and comes back. And by the way, this is not just true about product work. In my head, I was thinking about product stuff where he was like, “Hey, I think this product is doing this wrong.” He’s like, “No, no, that’s why it’s not that way.” And the product will start to shift. Also to give him feedback just on his own presence in a meeting or delivery, he’ll be like, “Oh, well here’s why I did it that way.” And then a couple weeks later, you’ll get in a similar situation, and he will moderate how he shows up. So I actually find him somebody who’s really satisfying to give him feedback. It really works. It’s very effective, but you do have to take the long view on it. And he will have a… The things he did, he didn’t do an accident. He will have a reason why he did them the way he did them.
The HPM Weekly Report
Lenny: It’s a great example of strong opinions, loosely held.
Andrew ‘Boz’ Bosworth (00:46:42): Yeah, that’s right.
It also makes me think, I think you used the compiler analogy. I’m thinking the model training. He’s retrained his model overnight.
Andrew ‘Boz’ Bosworth (00:46:50): Yeah, that’s right. It’s funny. One of the things that’s so funny about Mark is if you give him some feedback in the morning, the next six meetings he has, whether it’s about that product or not, he’ll ask people what they think of that feedback. He won’t attribute. He’s just like, “Hey, what do you think about this in this product? And so you’ll be in a meeting with him and you’ll see him doing it. He’ll come to the meeting with you about some other topic. He’ll be like, “Hey boss, what do you think about this product and this idea?” And so he will, over the course of the day, take that little note and kind of pressure test it, and he loves to triangulate. Where are all the points of view on this that maybe you didn’t see? So he really values a broad perspective on each thing that’s being discussed, which is really fun.
Trying to get more training data for his model. I get it.
Andrew ‘Boz’ Bosworth (00:47:32): You can’t get me to call Mark an LLM. That’s not fair. That’s [inaudible 00:47:35].
We could all hope to be as smart as Mark. As you were talking, I noticed your tattoos, and it reminded me that you’ve got at least two tattoos that I’m aware of.
Andrew ‘Boz’ Bosworth (00:47:35): Oh, yeah.
One is of California, which I completely understand. California is a very special place, but you have this other tattoo that is just the words Veritas. Can you talk about what that’s about and why that’s important to you?
Andrew ‘Boz’ Bosworth (00:47:58): The funny thing about tattoos in general is I came out of high school, I was like… I don’t know if there was an archetype for me, but I didn’t drink until I was 21. I was a very rule-following person. I was like, “Why are you going to get a tattoo, affect your body? “Why would you dye your hair? Just let it be what it was.” And some of this was I think I was somebody who was privileged and had a great deal of self-confidence in who I was and what I wanted to be, and that was fine. But in some ways I was also weirdly judgey about other people in a way that’s kind of off- brain for me, certainly today it would be, but at the time, getting a tattoo is a big deal for me. I was like, “Oh, this is just the vehicle for my life and you can do whatever you want with it. And it’s not like a… It’s something that you possess, and if you feel like if you want to decorate it, you can decorate it.
And so getting a tattoo was a big deal to me actually, and I completely shifted my mindset of how I thought about my body and how I thought about people’s body and their presence and their time, maybe to some degree, even an understanding of mortality. Like, “Hey, can’t take it with you. It’s all going to go…” When you’re 18, you think you’re going to live forever, and by the time you’re 22, a grizzled 22-year-old veteran, you’re like, “Ah, tattoo that bad boy up. It’s all going down.” And so yeah, that’s why I got the Veritas tattoo, which is Latin for truth, which is… I will say it’s a little cheesy, because it’s also Harvard’s motto. But I got it in a monotype font. This is the programmer’s font here.
The other thing that’s interesting to me about tattoos was it’s also part of a generational shift. We grew up in a time when tattoos were really seen by adults as gangs or bikers or sailors or certain types. Now my understanding, I saw a stat recently that more people in my generation have tattoos than don’t have tattoos. And so I think we also just culturally shifted positions in a way that… I find richness of self-expression wonderful. I really think it’s great. And so I’m here for all of it.
My assumption from what you’re describing is, this idea of truth is very important to the way you think and work.
Andrew ‘Boz’ Bosworth (00:50:01): My reputation does precede me on this point, I’m afraid, which is… I think when I was young, I saw being honest… And I was wrong by the way. I saw being honest as kind of a get out of jail free card. You could say whatever you wanted as long as you’re being honest. That’s just not the case at all. I’ve written about this before, but by far my biggest professional regrets were me not being kind. And I used to think… I wrote this note a while back called Be Kind, where being nice, that’s like patronizing, or telling somebody things that are half-truths or just getting by. And I’m against that. But being kind isn’t that. Being kind is like, “Hey, how can I deliver this feedback in a way that is actually productive and helpful, in a way that is going to help them and not cause them just to feel bad and helpless?
And I think I did that wrong a lot as a kid, as a young man. And so being honest is still a big part of my personality. No one would ever accuse me of being dishonest who knows me. And I think people understand and respect that I’m pretty direct, and if I have concerns or issues, I’m going to bring them up. I’m just much better at bringing them up now and expressing a true care and belief. I wouldn’t bring it up if I didn’t think we could do better, if I didn’t think we could fix it, if I didn’t believe in this situation. And so being honest is still a huge part of my identity, and I think that’s something I’m very proud of. But I will say the contextualization of how I’m honest has changed immensely since I got this tattoo.
That seems reasonable. This touches a little bit on something I definitely wanted to talk about, which is one of your most classic pieces, and this is the way I first learned about you, is a piece that is called Communication is the Job.
Andrew ‘Boz’ Bosworth (00:51:40): Yep.
I know many people have read this, many people haven’t. I’d love for you to just talk about what this means and why this is important, why this is something that you wanted to share.
Andrew ‘Boz’ Bosworth (00:51:49): Yeah, it’s one of the things that, especially if you aspire to be a leader… And leadership isn’t management, and leadership isn’t being the only person responsible, it’s not even always the same as accountability. But if you want to have an impact…
Andrew ‘Boz’ Bosworth (00:52:00): It’s not even always the same as accountability, but if you want to have an impact on the world around you, it is exclusively done through the creation of artifacts or verbalizations that affect other humans. That is all there is. That’s all there is if you want to have an impact, if you want to create some kind of a lasting change. And it could be in your little relationship, it could be in your team, it could be in your company, it could be in the world. It is down to communication. And so often you hear people saying, “Oh yeah, I wrote that up a year ago.” It’s like, “Yeah, but you did a bad job of writing it up a year ago, or we would’ve not wasted a year not doing it.”
People always think that’s, “Oh, I had that idea,” and that’s means anything. It means nothing. It means absolutely nothing. Or it’s like, “Oh, I wrote this post.” Well, you didn’t break through with it. That’s on you. It’s not on the audience. People want to blame the audience. Well, the audience is just there. I mentioned this even earlier, and I hope people caught it, when I said, “Hey,” I give somebody a piece of work and they come back six months later and they have done the wrong thing, I’ll take the L. I will take the L on it. It’s not great for them. They’ll be pissed they wasted their time, but like I said, that’s my responsibility. I did not communicate clearly what I wanted, what the expectations were.
Could they have also helped themselves? Sure, they could have, and that’s a thing that takes all sides. We should work on this problem for both angles. I have another post called Listening is the Job, which is the other side of this. But communication is the job is, I really believe… Actually, it’s the relationship to this idea that came out of the US Marines and the SEALs of extreme ownership, which is… So whenever thing goes wrong, I ask myself, “What could I have done differently in terms of how I communicated things for this to have gone better?” Could I have set priorities better? Could I have set expectations better? Did I need to have a better metric that I pointed the team at? Did I put the wrong people on it?
By the way, the thing I talk about is org charts are communication devices. They don’t exist. There’s not a physical string between you and your manager. They’re just communication tools that are supposed to give people a rough sense of how things are organized and where to go with who. And so all these things are communication. Silence is communication, me not reaching out to you to check on you, to check on your project. We talked about the [inaudible 00:54:29] earlier. What does that mean? That means trust. That means responsibility. The absence of check-ins has meaning. You cannot not communicate. You are always communicating something with your face, with your clothes, with your body. What are you communicating?
I’ll give you a funny example, which I hope we get to put in the podcast, because if you’re watching this on video, you will have noticed that my camera cannot stop adjusting light. It’s just constantly too dark or too bright. I’m trying a new camera. I’m a nerdy guy. I try a lot of camera gear, I try a lot of microphone gear. I love to have all the latest gadgets and gizmos, so I’m trying something new. It’s not working. And in my head I’m like, what is this communicating about me? People are going to think that I don’t care or that I’m not competent.
That’s what I’m talking about. And now, I felt compelled to explain it in the podcast so that I can communicate clearly that that’s not the case. I really just think so much of what I try to do in my professional life is understand the mental model of other people. Where are they right now? And I mean specific people, like my managers or my key technical leaders, and I mean general people like teams, and I mean broadly just the average human. Where are they at in this conversation? And how can I craft my language, my presence, my persona, everything to usher them from where they are to where I want to get them?
And that requires me to have a very clear idea of where I want to get them, have to have a clear idea of where they are. And I want to tell you, it’s not as much work as it sounds like. I think no one would accuse me of having this big fabricated persona. It’s not that. But it is having tremendous empathy for where people are starting. That was the lead for me. All the rest of it, all the rest of how I show up in meetings trying to smile more, because I’m like a big scary guy, those things are little things that you work on and they become second nature and they’re easy. The hard thing is just having the empathy for your audience and being, “Where are they? Where are they starting?” And when you miss taking responsibility for that, extreme responsibility for that.
Information Transparency and Team Trust
Lenny: There’s so much good advice in that. There’s so many threads I want to follow, but let’s just follow this last one of trying to understand how someone is best communicated to. Is there an example to make that a little more real for people of just what you’ve done to, “Oh, here’s how I’m going to communicate with this person?”
Andrew ‘Boz’ Bosworth (00:56:46): I’ll give you a couple. One is multimodality. There’s an old saying, repetition never spoiled the prayer. And I think most experienced communicators, whether they be writers, whether they be public speakers, talk about the importance of reiterating a point several times and in several different ways to make sure that people have a chance to internalize it. You want to use it, you want to say it directly, you want to use metaphor. And so for me, I will give an all-hands and then write a post with the content of the all-hands, because different people are going to respond differently to these modalities and are going to absorb information at different rates on these different modalities. That’s a trivial one.
Another one that I think of all the time is making sure that you address people’s fears and concerns. People will not listen to you if they think you don’t know what’s going on. And so, one of my favorite things to do when we’re talking about some kind of issue is right up top and say, “Hey, let me be clear. This is the issue we’re having. I know we’re having it. I know what matters.” And then I’ll say the same thing that I would’ve said, but they would’ve literally ignored me, because how can they trust my conclusions if they don’t accept the premise? You know what I’m saying? I think there’s a whole piece there. Obviously when you’re in person, it’s a lot easier because you’re reading facial expressions. Even on this, I’m reading you nodding on that. I’m like, “Okay, he’s with me.” And then I throw in a, “You know what I mean?” Whereas if you give me a cocked head, I then bring a second example to try to drive the point home. But you build yourself up. Most people are going to realistically start in their careers trying to influence one or two people. That’s where you start. One or two people. That’s who you got to communicate with. Your manager, one teammate, that’s who you got. And then you build up and build up and build up a skillset to do it at a larger and larger scale.
I love so much of this advice. I think it’s also helpful for relationships. “Here’s what you’re upset about.”
Andrew ‘Boz’ Bosworth (00:58:42): Totally, a hundred percent.
“Here’s what I think we can do.”
Andrew ‘Boz’ Bosworth (00:58:46): Again, the work that I’ve had so graciously supported to do on myself at Meta with great mentors, Sheryl Sandberg, Mark Zuckerberg, a bunch of others and coaches absolutely made me a better partner and husband, and my wife. And then by the way, vice versa. Having kids and getting deep in the literature around raising children. Congratulations to you, by the way. Getting deep in that literature made me a better manager. Absolutely made me a better manager in terms of thinking about how people are managing their emotions and how to engage with them in those times.
Top-Down vs. Bottom-Up
Lenny: Amazing. We need a second edition of this Boz’s Parenting Advice.
Andrew ‘Boz’ Bosworth (00:59:19): That’s right.
And relationship advice.
Andrew ‘Boz’ Bosworth (00:59:21): It’s all the good stuff. No bad kids, Lansbury, it’s good inside, Dr. Becky. It’s all-
I like Dr. Becky.
Andrew ‘Boz’ Bosworth (00:59:28): I really think that the modern parenting canon is really rich.
As a product leader, you probably have these requests on your roadmap, but the struggle to prioritize them is real. Building analytics from scratch can be time-consuming, expensive, and a really challenging process. Enter Explo. Explo is a fully white-labeled embedded analytics solution designed entirely with your user in mind. Getting started is easy. Explo connects to any relational database or warehouse, and with its low-code functionality, you can build and style dashboards in minutes.
Once you’re ready, simply embed the dashboard or report into your application with a tiny code snippet. The best part, your end-users can use Explo’s AI features for their own report and dashboard generation, eliminating customer data requests for your support team. Build and embed a fully white-labeled analytics experience in days. Try it for free at explo.co/lenny. That’s E-X-P-L-O dot C-O slash Lenny.
Okay. You mentioned this gadget and these cameras you like to play with. Let’s talk about the Vision Pro and VR headsets. Have you tried the Vision Pro? Thoughts?
Andrew ‘Boz’ Bosworth (01:00:57): Actually, Mark and I tried it together and I want to say, first of all, when the headset came out, we breathed a little sigh of relief because the stuff inside of it didn’t represent a fundamental breakthrough. Everything inside of it was something that we probably could have gone and done, with the exception of Apple silicon, which is a marvel, but it’s worth nothing. Apple Silicon is like a 2X marvel when doing things like scaling display resolution is unfortunately a quadratic proposition, and so a 2X linear scaling advantage doesn’t buy you as much as you might expect when you’re trying to scale resolution. And so that was step one. But we still assumed at that price point, with their legendary attention to detail and polish, they probably produced a great product. And the line that I said actually my own podcast with Matthew Ball a week ago was, “Look, I was prepared to come to market and say we have the best value headset. If you want an outstanding, best possible headset for the money, we’ve got it. It’s the Quest 3.”
And I was so thrilled when I tried the AVP next to Mark, we were like, “No, we think we have the actual best headset.” Now, we’re not saying it’s the best at all things. If you’re sitting still and watching a movie a high-res movie, Apple Vision Pro is really great. It’s really great. The resolution shines. The way they’ve tuned the pass-through shines if you’re stationary and looking straight ahead. And they’ve done some really nice things with the UI. It’s one of these things that we do get annoyed about, a mile aside, we get a little bit annoyed about as product people.
I know that this happens to all of us. It happens to Apple, happens to Google, happens to us. We have a bunch of internal things we’ve been playing with, which will at some point ship and we will be accused of having stolen them when we actually did not steal them. If you want, you can go see my Quora answer on the history of the like button where this happened previously, where we had built the like button internally before it was launched elsewhere. Anyways, it’s a whole thing. This happens in our industry a lot, and I really shouldn’t care as much. It’s a little bit of my ego peeking through, which I should control and tamp down, if I’m being responsible.
The beautiful UI polishes, they did a tremendous job with eye tracking. One of the things that’s interesting about the eye tracking is, to do it the way they’ve done it, that’s why you have to have the prescription insert, so it doesn’t support your glasses. You have to get prescription inserts. They’re expensive. And they can shoot the cameras that track your eyes through the lens as well as the light around it. Ours go from the side on the Quest Pro, and that allows you to wear corrective lenses.
And so different choices like that have trade-offs, but it’s still cool. It’s great that they got that in there. At the same time, our hand tracking is better. Obviously, the app library we knew was going to be better. That’s not totally fair to them. They’ve just launched and they have small volume still, but I just find the comfort… The thing that really got me the most, the field of view is really small on the Apple Vision Pro. And some people are characterizing it incorrectly on the internet. They’re doing a characterization up close to the lens. Once you factor in the eye relief, the distance between where the lenses are and your eyeball is, their field of view gets pretty narrow for almost all faces relative to ours, which I find distracting.
Their displays are much dimmer than ours, and I find the motion blur really distracting when I’m in mixed reality use cases. And as I mentioned earlier in the podcast, I’m a huge mixed reality buff. I’m a huge fan of that potential for exactly the same reason that they are, by the way, which I think hands and mixed reality make it feel much more accessible to more people. I’m pretty glad we have the controller in our set though, because it really expands what you can do. And we don’t just operate our computers with just one thing. We have a keyboard, we have a mouse, we do multiple modalities all the time. I really feel like the comfort, the lack of persistence and motion blur in our pass-through, the brightness of our displays, I was like, “Oh man, if you gave me one to take, I would take Quest 3.”
Now, people have rightly said, “That’s pretty biased.” Of course it is. Go get your own opinion. But what kills me is most people haven’t done that. They have not tried the Quest 3. That’s what kills me the most. If you go and try Quest 3, ask yourself if you’d rather have seven of those, one for you and six of your best friends, or one Apple Vision Pro. I’m sure the answer isn’t Quest 3 for every person. There are people for whom there are use cases that really fit their life, the Apple Vision Pro, I’m cool with it.
But people don’t even know that the Quest 3, you can do Remote Desktop. You can do it both through an app called Remote Desktop, which is very popular, or you can go into workrooms and you can have three monitors surrounding you streamed from your machine. I think some of this is just people have not even done the work. They haven’t even tried it. I welcome all of you who think I’m biased to prove one way or the other what you think, but don’t do it without putting the Quest 3 on and giving it through its paces, because it’s a pretty great device. And you can do a lot with $3,000 extra dollars.
How did they get away with that, by the way? We launched a headset that was 3,500, it’s fine. This is fine. No one cares.” I don’t know. Fairness is too much to ask, and I don’t care about that. Apple has earned the great brand that they’ve built. They truly have. I think it’s tremendous. I certainly celebrate a large number of Apple product. I’m a huge fan of their work. I’m a huge fan of what they do. That’s probably why I expected more from the AVP.
Clear Boundaries and Guardrails
Lenny: Well, I’ll show you my favorite AR device, which is these Ray-Bans. I actually bought them.
Andrew ‘Boz’ Bosworth (01:06:13): Yeah.
Here, I’ll put them on it. Here, what I’m going to do is I’m going to record. I didn’t tell you I was going to do this. I’m going to record this as we’re talking. Look at this.
Andrew ‘Boz’ Bosworth (01:06:20): I like that. They look so good on you too. This is a good fit.
My mother-in-law’s like, “You look so sophisticated. You look really smart with these on.” And we bought these actually to film our kids, or our kid, instead of having Cameron’s face. And it’s been awesome.
Andrew ‘Boz’ Bosworth (01:06:34): It’s really the best. It is so hard to look at a phone screen and have the real thing be in between you. No, the glass is the way to do it.
Look at this, my new look. I’m just going to have glasses on all the time.
Andrew ‘Boz’ Bosworth (01:06:46): We got to get you the multimodal. I’ve been playing with that since December, where you can use the camera to ask a Meta AI assistant about things. It’s really good. I was in a ski village recently with my family and I had them on. I’m just like, “Hey, take a look. Tell what you see.” And I had found a sign and it gave me directions. “Hey, the bathrooms are down those stairs to the right. Do you want food? It’s over to the right.” It couldn’t tell what village I was in, but it was like, “You’re in a ski village somewhere. Here’s the amenities.” And I was like, “Wow.” There’s something real magical here.
I feel like I need this for my podcasting interview so I could just have a little voice tell me questions to be asking and where I’m at.
Andrew ‘Boz’ Bosworth (01:07:25): Right, totally. Yeah.
How helpful would that be?
Andrew ‘Boz’ Bosworth (01:07:27): Okay, I’m going to cheat on this one and say, obviously we’ve been talking for a while, but playing with glasses that have full AR capabilities, and we’ve got one that has rumored to be coming internally soon, so heavily rumored in fact that you might even say it’s almost been confirmed. And what’s been really fun is being able to play with these time machines, really, in terms of what they are. It’s amazing technology. People were giving us speeches, big company-wide speeches, and had all their notes on the glasses. And they could control the slides just using a gesture.
Oh my God.
Andrew ‘Boz’ Bosworth (01:08:07): There’s exciting things afoot. The future’s looking pretty bright.
Something that I wanted to touch on, which is when Mark put out this whole video of, “Here’s what I think of Apple Vision Pro versus the Quest,” a lot of people are just like, “Oh man, because he’s putting out this video, he must be so afraid of what’s happening and it’s not the right move.” What went into thinking? Was there strategic thinking there? Was it just him, “Here’s what I think?”
Andrew ‘Boz’ Bosworth (01:08:29): This guy does [inaudible 01:08:30] about the modern era. Everyone’s in their Meta head about everything. Everything’s a four-dimensional chess. That was just what Mark thought of the thing. That’s what he thought about it. I think he wanted to make sure people remembered, hey, Quest 3 is literally a better device, and people haven’t even tried it.
And so we’re not always playing four-dimensional chess over here. Sometimes we’re just like, “Here’s a thing that I believe is true. I’m going to say it out loud with my mouth.” That’s what he did. It’s not that hard. When I do it, everyone expects it because of my persona or brand or whatever the thing is. I guess people are surprised when a CEO does it. All right, I get that. That’s cool. There’s other sets of societal expectations there.
And we’re all familiar with Apple putting the Welcome IBM ad in the New York Times, and then Slack doing the same thing with Microsoft and the Ballmer iPhone comments. None of those were true discussions of the technical merits of a product. Those were all just big, rally the troops gestures. This is not that. Mark is deep. He’s an expert in this stuff. I’m an expert in this stuff. I feel great about our choices.
By the way, when I’ve used it, I could get myself completely into the head of the person who designed it. I can tell you from using it what instructions they were given, that team was given. I can tell you what they were optimizing for. I can tell you what constraints they were under. By all the choices they made, I can tell you all those things and I understand it coheres in that way. We made different choices. It shouldn’t surprise anyone. We like our choices better. We could have made those choices. We didn’t make those choices. We made these other choices.
And so for me, the weight, the wire, which just always brushes against my ear, the pocketable thing, I get it. That’s not what I would’ve done, and I know that because I had the chance to do it and I chose not to. I don’t know. I don’t know why people are surprised. This wasn’t some big, savvy strategic move. This was just, Mark’s got a chance to use it. He’s like, “Oh man, I think we should tell people what the real story is here.” And we did.
I love that insight. And I know a lot of people watching are like, “Oh, shit, he’s right. Wow. I didn’t think of it that way.” And so I think it had a lot of that impact.
I want to zoom out a little bit and talk about Meta’s journey over the past couple years. It feels like there is a huge downturn in public perception of Meta and the stock price, and then over the past couple of years there’s been a huge turnaround. And it feels like there’s always a lot to learn from these periods. Just as an example of the stock price, I was just looking at it, it was down to 487.
Andrew ‘Boz’ Bosworth (01:11:05): Yeah.
I’m curious just what you’ve learned from going through that downturn and turnaround. And I know it’s still in progress, but just what have you learned from that journey?
Andrew ‘Boz’ Bosworth (01:11:16): Yeah. Well, there’s a lot to take away, I got to tell you. I think we had the largest single day stock drop in history, followed 18 months later by this largest single day gain in stock market history.
Oh my God.
Andrew ‘Boz’ Bosworth (01:11:27): As the legendary Lou Holtz said, ” You’re never as good as they say you are when you’re winning, and you’re never as bad as they say you are when you’re losing.” Mark has always brought that quote out to guide us internally to try to insulate ourselves a little bit from the vagaries of external opinion. And that’s not just true with stock prices. That’s true with media and press. It’s true across a lot of things.
One of the things I told my team, and I still have to repeat it to them, is, “One of the things that’s hard to remember when you’re in it is that you know more than the critics do. You know more than the analysts in the marketplace do. You know more than the media does. You know more than the podcasters do. You know more than the Twitter does. You know more about what’s real and substantial of value about our company than they do. That doesn’t mean ignore them, because they have a different perspective and you need to understand it. Even if the totality is less than what you know, it may contain parts that you do not know.” I’m a huge fan, I read the criticism of everything, and I read it very carefully, looking out for confirmation bias, looking out for things that I might be inclined to resist but are maybe true. I invite all critique, but I also don’t accept the critique blindly. I don’t just say, “Yes, this is obviously true.”
Gell-Mann Amnesia is a great concept for everyone to understand. Gell-Mann Amnesia is this property where you’ll read a newspaper article, let’s say newspaper, why not, about a thing about which you are an expert, and you’ll be baffled because here is an article that is not just wrong, it’s inverted causality. Michael Crichton, I’ll steal Michael Crichton’s quote on this, “It’s a wet sidewalks make rain story.” And you’ll be like, “What a terrible, bizarre story.”
And then you will turn the page of the newspaper and it will be another article about a topic about which you know nothing, and you will read it as if it is the gospel truth. You’ll figure it’s, “Oh, no, look at this information about this foreign situation.” Like I said, perfectly true. We should be smarter than that. And so does that mean you don’t read the thing? No, you read it. You just read everything with that perspective of, “Wait a second, this is another point of view. And how do I integrate that into a whole perspective that I can have and be informed about?”
The first, the macro thing is taking the long view, realizing that when you’re in the dumps, it’s not as bad as you think, when you’re at the top, it’s not as good as you think. It’s somewhere in between at all times. The second thing is communication is the job. We really did not communicate effectively, I think, with the market around our future investments. And listen, we’ve had two 10-year long huge investment areas. One has been AI, one has been reality labs. And AI’s looking pretty good today. I can’t think we can all agree with Lama 2, with Fair, the breakthroughs that we’ve had.
People don’t notice that Fair, our AI research lab, is the second most cited research lab in AI behind Google. We’ve been doing this work. We didn’t come here casually. We’ve been doing it, and so that’s looking pretty good. I don’t think we did enough to explain those best to people. Previously, the core business was going strong enough and they were willing to ignore them. And with the old Warren Buffett quote, “It’s only when the tide goes out, we see who’s not wearing swimming trunks.” And so when the tide went out, when you have the Ukraine war and an interest rate hike and recessions, now everyone’s scrambling for that incremental dollar and they’re like, “Go get rid of this stuff.”
And we had to tell the company, “You don’t want to work at a company that, when times are tough, kills all future growth and just shores up in the core business.” That’s a company that’s just committing itself to dying at some point, a little later than expected, but dying at some point. You want to work at a company that has a balanced portfolio of investments, which we had. We didn’t explain that well, and so we spent that time explaining that to the market, to the press, to everybody.
And now, I think as people understand the size of it and the scope of it, and of course it helps that the core business has overcome its challenges there from ATT and the other kinds of stuff, it’s looking pretty good. I do think one part is, as an internal person, really moderating your attachment to the external narratives and swings, that’s super important. And you do that based on understanding your own expertise. And the second part is understanding why is there a delta. What is there? And grab that. It’s usually communication.
How Deep Should Leaders Dive
Lenny: There’s also a big flattening of the org. This was something a lot of people talked about, where managers became ICs. Is there anything more there that you’ve learned of just how to adjust the org to be more efficient?
Andrew ‘Boz’ Bosworth (01:16:22): Of course, and I should have included that in the first section. I was a bit eager to wrap it up elegantly in the two. But you’re right.
I appreciate it.
Andrew ‘Boz’ Bosworth (01:16:29): We made significant shifts in how we operated the business, which was super painful. And listen, this goes back to the boom times of COVID, when it looked like there was a real lasting secular shift in things like e-commerce and in working remotely and these tool sits, which are exactly what we build, and it’s primed us. And so we built up a huge workforce to pursue those opportunities. We still believe in those opportunities, but they’re back on their original timeline. And actually, literally, if you look at a bunch of graphs that we have internally, it’s literally the COVID boom and then, not bust really, but as it receded, everything’s back on its original trajectory.
We didn’t lose ground or lose time, but the pull forward didn’t happen. Well, that means your economics don’t make any sense anymore. Now, you made a bunch of investments that are going to yield too distant in the future, and getting there faster isn’t going to help you and you’re carrying a much extra cost. That sucks, man. It sucks. And we don’t feel great about it. We really don’t. It’s a business, it’s awful, and it happens.
I do think one thing that was interesting about that time was, for those of us who grew up and saw the .com boom firsthand, I was born and raised in the valley, so that was all around me when I was graduating high school and going to college, and then in the 2008 major recession on the housing crash and on the market and all that stuff. Now, let’s imagine you graduated college in 2009-
Andrew ‘Boz’ Bosworth (01:18:00): … all that stuff. Now let’s imagine you graduated college in 2009 and got a job. Well shoot, you’re 15 years in your career, you could be a director and you’ve never seen a downturn. I think we also had, in addition to what was very unfortunately conventional mis-forecasting in the business that caused us to over hire, that we had to correct for. You also had a workforce that was just not at all of a mentality that this could ever happen. This felt like it was an act of God when in fact it’s like a cyclical nature of all businesses that this will happen at some time and you hope it doesn’t and you wish it didn’t, but you have to deal with it. And so I think we have a bit of a tough storm there for the whole industry and we’re still feeling it. I think we’re still feeling it.
Certainly we’re happy at Meta to be beyond that point and we’re growing again and executing at a stable rate and feeling really good about that, but quite a few of our other companies in the industry. And it’s a very uncertain time for engineers, for PMs, for designers, for everyone and all the support functions around them. I’m super sympathetic for that. I think obviously the mis-forecasting that happened inside of Meta’s walls happened everywhere and now that you have, especially with higher interest rates and cash isn’t as cheap, runways are tighter. People are just making those pragmatic calls. I think we’ll rally back from this. I think this is a normal thing that happens to industries, but it doesn’t reduce my sympathy and empathy for those who have been affected by it or who live in fear of it.
The Eye of Sauron
Lenny: I was talking to a friend who works at Meta and I was asking them what it’s like to work at Meta and she was just like, “It’s intense, and it used to be more chill. There were people that were coasting here and there,” and now she’s like, “No, all those people are gone now. It’s just only the intense people left and we’re working really hard.” Does that bring up anything?
Andrew ‘Boz’ Bosworth (01:19:48): Yeah. I don’t want to comment on people who left. People left Meta for all kinds of different reasons and likewise role elimination happened in many cases because we just decided not to do this work at this point. We were going to do it two years from now and don’t need to carry a team to do it. I think it’s really hard to generalize because each of these is a specific person with a specific life and a story that is rich and deserves to be told. But I do think that if there was somebody coasting and you as a manager have to make tough calls on who you’re bringing in which way you’re going to bias my profound suspicion, and again, I don’t know your friend who you spoke with, my profound suspicion is that person’s probably already working hard. You know what I’m saying? That person’s already probably working hard. I don’t think we changed how hard any individual worked. I really don’t believe that.
I do think there was a selection bias as to what was going to happen, and I think that’s probably what you saw play. In fact, if there is a generality that can be found.
Maybe as a last question, I have this segment where I call Failure Corner where I ask people to share a failure of their career and what they learned from that experience. Is there something that comes to mind?
Andrew ‘Boz’ Bosworth (01:20:55): I’ve failed tons of times. I’ve built products that nobody used. I’ve built technical architectures that didn’t scale. I’ve failed all kinds of times. I don’t regret most of those. Almost every one of those I learned from. It was a stop on the path to a better solution or it was a recognition that this thing wasn’t going to work ever, which is its own kind of a gift. All the failures that I regret, that I take seriously are personal failures where I affected a person in a way that I’m not proud of, maybe wasn’t proud of the time because I wasn’t in control of my own emotions or mood. I was feeling fearful, I was feeling scared, and there’s a bunch of these one or two that stand out that I don’t feel comfortable sharing because the person affected, I think would prefer I didn’t share.
I’ll share one that was… I think the person’s and I are tight now. We have this really silly discussion. I remember it so vividly. In the early days of client server architectures, which Facebook obviously is a website, so you’re calling to a server to get the webpage, but then that server is going to call the other servers to gather things and that was one of the major clients of remote procedure calls because newsfeed ranking was all done on this other set of servers that had its own special requirements and special build and how it was put together. And so your main web server would put a call-out over a remote procedure, call to the remote server and get a response back. And we had this really janky RPC system that, I won’t say who built it, but it was built and it was just a piece of garbage constantly failing. It was not robust at all. And one of our best engineers, Mark Sleek, built a new one called Thrift. It was a great, really great RPC infrastructure.
And one of my best friends, Dave Federman, one of my really good friends and a brilliant engineer, we were talking about how to do the encoding and I was like, “I want it to be binary encoding.” I was like, “You should binary encoding. I want it to be super efficient on the wire,” because I’m storing these RPCs. They do two jobs for us, one of which is the active RPC, but I also store the RPCs in a log and replay them to do the work that we were doing in newsfeed ranking. That’s how it was done back in the day. It was all kind of asynchronous offline and so I wanted to be as tight as possible. My memory bandwidth was very limited and memory was so expensive back then and Dave Federman was like, “No, no, no, that’s short-term thinking Boz, we should be using Linux style descriptors that are plain English language. Then you could look at the log, you can see what it is. It’s possible memory bandwidth will get cheaper, but these logs being scrutable to development is going to be a better thing.” This is nerd bait.
Those of you who have been engineers in this call, this is like Vim EMX. This is nerd bait. This goes deep. This is like a long old thing.
I love it. Keep it coming.
Andrew ‘Boz’ Bosworth (01:23:52): It’s a room full of engineers at the company and the company’s not that big, and so there’s probably half the engineers of the company in this room. I was yelling. Literally I’m turning red, I’m sweating. I’m so angry at Dave Federman for countermanding my proposal and I’m the major RPC customer. A couple things. This is so dumb. Mark Sleek just built two encoders. It’s not that hard. You pick which encoder you want for your things. That’s the easiest solution ever. Second thing is Dave was right, by the way. Within a year, the memory bandwidth definitely didn’t matter relative to how inscrutable it was to try to get into these logs. I had to build a ton of extra custom software to parse the logs and understand what what’s going on. But also it was a case where my identity was caught up in being right. And for those who don’t know, identity threat is just the biggest… Your worst behavior is always going to come out when you think you’re under identity threat.
When you feel like some core part of how you see yourself is in question, you will react with every ounce of your fiber to defend that conception of yourself because it’s so expensive to reconceptualize who you are that you defend yourself. So my identity was being right. [inaudible 01:25:08] on the wrist. And it caused me to take one of my best friends, one of the best engineers I knew, a guy I literally lived with and getting a really embarrassing for me conflict, which everyone was just scratching their head like, “What is going on with Boz right now?” I looked like an unhinged crazy person. I remember the room, I remember where I was standing in the room. I remember everything about that moment and I had to go home and be like, “What the fuck was that? What happened?” I’m asking myself like, “What happened there?” And that was one of the many steps in the journey to recontextualizing what it was was not to be right and what meant to be open-minded and curious and how to engage in this competition.
I was 22. I don’t make excuses for it now, but I remember there was a couple other examples like that in things that were less technical and more personal that I won’t share, but I remember each of them vividly and those are the real failures for me.
The News Feed’s Birth and Controversy
Lenny: I love how this story is so long ago at this point and it’s still stuck with you and such an impact, I think.
Andrew ‘Boz’ Bosworth (01:26:06): Oh, my god, I’ll never forget that. It was embarrassing. And for those, that’s the old quote, it’s really one of the truest quotes, and I know it’s cliche and sometimes cliches are cliches because they’re good. It’s like, “People, they don’t remember what you said. They just remember how you made them feel.” That’s all anyone remembers is how you made them feel, and I think in that room, I made people feel unsafe, maybe. It was bad.
I like this concept of identity threat. They call this podcast episode identity threat.
Andrew ‘Boz’ Bosworth (01:26:32): There you go.
Boz, we started this episode with a billion questions. I have a billion and one questions now. I wish we could keep going, but I know we have to wrap this up. Is there anything you wanted to share or leave listeners with before we get to our very exciting lightning round?
Andrew ‘Boz’ Bosworth (01:26:46): Well, in the off chance that we do end up labeling this episode, identity threat, let me give you what really was… I made a lot of breakthroughs with coaching, learning about the feeling inside of my own body when I was feeling that identity threat and learning techniques and tactics to reduce the likelihood I would feel it and how to deal with it when it happened and how to repair when I did, I went through all that stuff. I would say the greatest lesson I learned would come years later and it was just from observing somebody. Ami Vora who was a legendary, long time came in and worked on our development platform, then worked with me on ads for a long time and then was the PM lead for WhatsApp for a long time. She’s since gone on to do even more great things outside the company. Working with her, it was like watching an alien because her and I were so different in our approach and she approached…
She could have the most profound disagreement with somebody in the world and they would say the thing that she thought was not just wrong, but crazy wrong and she would respond. She would say, “Fascinating. You have to tell me more about why you think that.” And I can’t do it justice, she meant it from the core of her being. She saw this schism between her and that person and it could have been personal, it could have been professional, it could have been anything. She saw this schism and how they saw the world and how she saw the world and rather than reacting as if it was a threat that somebody saw it differently or rather than reacting afraid that maybe she was wrong and had done things wrong before, she reacted with the most genuine and profound curiosity. I just watched it absolutely tear down walls between points of view. People felt immediately her genuine heartfelt curiosity and would lean in and that would cause them to be open-minded.
And if she was right, which by the way she usually was, then they would leave being like, “Oh, okay, I was wrong about that,” but she also would change her mind and that was the key. Ever since then, I really have tried to model that. When I have a strong internal clinch, I try to embrace curiosity like, “Wow, we do not see this thing the same way. That is fascinating. Tell me why you see that.” And that can be by personal feedback. Someone’s like, “Hey boss, I don’t think you talk enough.” “Wow, you don’t think I talk enough? That is unexpected. I would love to hear more about that because no one’s ever said that to me,” so I wanted to give that to anybody who might recognize this behavior in themselves. There’s lots of things that you can do and you should do that work.
The work of improving yourself is always fruitful and satisfying and it pays off, as we discussed, in every aspect of your life with your family, with your friends, and professionally. This is one that I really thought was so great was just curiosity. Embracing curiosity in those moments of challenge has really completely changed my life and I owe that to Ami Vora.
Wow, I love this example. It’s basically an example of, “Yes, and?”, but in a really… no one’s going to be like, “Yes, and?” It’s a really nice way of saying it, “Just fascinating. Tell me more.”
Andrew ‘Boz’ Bosworth (01:29:50): Fascinating. You’ve got to tell me why, I want to understand it.
I love that. Maybe that’ll be the new title, fascinating.
Andrew ‘Boz’ Bosworth (01:29:56): There you go.
Anyway, with that, Boz, we’ve reached our very exciting lightning round. Are you ready?
Andrew ‘Boz’ Bosworth (01:30:03): I’m ready.
First question, what are two or three books that you’ve recommended most to other people?
Andrew ‘Boz’ Bosworth (01:30:09): The Dream Machine, which is a tremendous history of pre-history, really of modern computing ostensibly following the life of J.C.R Licklider, but really it’s much broader than that, is a tremendous missing piece of history in my opinion. I think in my discipline, in computer science, we were not properly educated. We learned about Alan Turing and we learned about some of the technical underpinnings of computer science theory, but the modern computer and the path to it is a profound and fascinating one, and it has particular resonance today as J.C.R Licklider’s observation was human in the loop computing and I think we are now in human in the loop AI, and I think there’s a tremendous resonance there. The second one is Good Inside, which is Dr. Becky’s book, and again, I think it’s a tremendous parenting book, but more than that, it does contain lessons for how we think about our own emotions and how we manage those, which I find to be useful in any context.
Amazing recommendations. She’s also got an online community for people that-
Andrew ‘Boz’ Bosworth (01:31:03): She has a wonderful online community, she’s very engaged in it. And for those parents out there, you’re a little early for this Lenny, but you’ll get there. Having little scripts that I’m reading on Instagram that when I’m in a moment of tremendous emotional challenge with my children, I have the words handy. They’re just top of mind for me. They’ve been cashed in, right? They’re primed, is a big game changer.
This will be for our parenting episode down the road.
Andrew ‘Boz’ Bosworth (01:31:27): That’s right.
Is there a favorite recent movie or TV show that you’ve really enjoyed?
Andrew ‘Boz’ Bosworth (01:31:32): Yeah. This is super conventional, but Mandalorian. One of the things that’s been fun about that is really my kids again is we watched it with my kids, so we had a chance to go on the Galactic Star Cruiser. Again, y’all know I am a genuine and true nerd through and through. Huge fan of Scott Trowbridge and his work at Star Wars service lands in Disney and also of that. And so before that we got our kids who were nine and six and watched all the movies together, and then we were watching Mandalorian together as a family, and it’s super fun and it’s fun to have that kind of lore, a connection. So it’s not just the classic kids movies, but there’s something more, and I think for them they feel like it’s an adult kind of conversation they get to be a part of. I’ve really enjoyed that. And I think the world of Dave Filoni and John Favreau and the team that’s building that universe out.
This is the way.
Andrew ‘Boz’ Bosworth (01:32:19): This is the way.
Do you have a favorite interview question that you like to ask candidates that you’re interviewing?
Andrew ‘Boz’ Bosworth (01:32:25): One of the most important things that I always ask people is what people who’ve worked with them would say are their greatest strengths and weaknesses. I like this for a couple of reasons, not least of which is I often do follow up with references and I like to triangulate their awareness of how other people calibrate though. And also how they respond to criticism. Sometimes candidates surprise me. They say, “Hey, you’ll hear this critique a lot from me. I don’t think it’s fair.” That can be an okay answer, but they’ve got to be a pretty robust there. Or it’s like, “Hey, this is something I’m working on, here’s what I’m working on.” But I also like to hear what they think their superpowers are. And too often a lot of attention in interviews is paid to weaknesses, which I care about because I want to know what the downside is. But way more important to me is like, “What are you awesome at? What is the thing that if I can just hitch a wagon and ride, that’s what I want to know. Where’s the superpower that you’re crushing it at?”
And what’s funny is people are pretty rarely give my reference checks. They’re not that often accurate about what the critiques are, but they’re usually pretty accurate about what their strengths are.
Passion and Career Exploration
Lenny: Interesting. I’m also a huge proponent of strengths and not worrying too much about your weaknesses.
Andrew ‘Boz’ Bosworth (01:33:36): But asking people to contextualize their performance through the team, that’s so important. We do not achieve very much on our own.
Communication is the job. Amazing. Is there a favorite product that you recently discovered that you love? And Meta products are acceptable answers.
Andrew ‘Boz’ Bosworth (01:33:55): The Ray-Ban Meta glasses would be a tempting answer. The multimodal stuff, which I’m going to get in trouble because I’ve been teasing this for months on social media and it’s still a closed beta, and it’s like, “We’re growing the beta slowly,” but it’s like, “People already really badly want it.” It was probably one of the most magical things I’ve gotten to try recently. It’s not totally fair because it’s also not you can buy the Ray-Ban Meta glasses and very soon, I won’t say when, but very soon they’ll have this capability, but it’s more fun to think outside of the box. All right, this is such a… I’m going to get in trouble for this answer because it’s a bit of a pretentious one. I’m not a car guy, let me say that right now. I don’t like cars. I don’t care about cars. I want a car that works and doesn’t break. All my cars growing up, had over a hundred thousand miles in them because we only ever had used cars.
They were constantly breaking down and the power steering would go out while you’re driving and you have to [inaudible 01:34:43] the thing and the brakes would go out or you’d throw a rod. I’ve done it all. I just wanted a car. And so then I drove a Honda Accord that I bought for 10 years and then I drove a Tesla Model S for 10 years. And a Tesla model S had a thing happen to it while it was parked, I will get into that. I get a new car and again, I’m like, “I want to get an electric car.” And I was like, “I’m going to get something nice. I’m going to get something nice. I want to get something nice.” I’m driving a Mercedes-Benz AMG EQS, and I didn’t know cars could be this nice. I grew up driving used cars and whatever, I did not know it was possible. It is the best augmented reality product I think you could buy in the market today.
With the heads of display, it puts your turn in three-dimensional space. It’s got cameras facing your eyes, so it’s positioned correctly on this display relative to your eyes, the turn. And so when you come up, it’s like you’re hitting a little wall, you got to turn before you hit the wall. And I was like, “Oh my God, I think this car has a great augmented reality.” Not a car guy, and I’m not trying to flex on this car or whatever it is. I like it a lot. I didn’t know they could be that nice, but the thing that I thought was so impressive was they did an amazing job with augmented reality in this car.
Wow. Mercedes-Benz, a player in the augmented reality, mixed reality space.
Andrew ‘Boz’ Bosworth (01:36:01): Big time. They’re out there, they’re in the lead.
I sometimes think about having a contest where I give away products people mention in this segment, and now you’ve blown my budget way out of proportion.
Andrew ‘Boz’ Bosworth (01:36:12): Ray-Ban Metas, you can afford that.
That’s an amazing, I’m going to have to check that out. I think we’re going to increase some sales for Mercedes. Next question. Do you have a favorite life motto that you often come back to think about, share with friends and family, find useful?
Andrew ‘Boz’ Bosworth (01:36:27): Yeah. We have a funny actually how we came about this. The motto my family has, my immediate family, me, my wife and kids is just trust yourself. Just trust yourself. Actually, the reason we have that motto is when we have a house, we have a few nice art pieces and one of them is a Tracy Amen, a famous UK based artist, and she does neon pieces and it says, “Trust yourself.” And so it’s in our bedroom hallway where me and all the kids and my wife are, it says, trust yourself and every morning the neon lights up and it’s trust yourself. And then I had the chance to have a crest made in the UK and my family’s English from way back. And so in the crest it says, “Trust yourself.” And I always talk to my kids, especially. I really think people, when you’re experiencing peer pressure like, “Who do you trust? Them or yourself? When you’re having a lot of self-doubt and uncertainty, you have to trust yourself.”
I just think so much of the success I’ve had, I think this is probably true of most people who went to startups and succeeded, was like, “I just had faith that I was making good decisions.” I mean, this comes back to the conviction point I made earlier about how you do things like newsfeed or controversial things or how you make big expensive changes, just conviction. You have to believe you, your eyes, ears in an intellect have combined to give you a point of view that has intrinsic value and deserves your respect as opposed to reading that newspaper article about your company and believing it over what your own eyes and ears have told you. That’s the motto that we go with.
And I think it’s also important to say you won’t always be right and that’s okay.
Andrew ‘Boz’ Bosworth (01:37:59): Totally. That’s right. Trusting yourself also includes taking risks because you trust that you can deal and handle with what happens when the risks don’t pay out.
Beautiful. Final question. I know that you’re an amateur photographer, maybe semi-pro photographer, a lot of travel photography.
Andrew ‘Boz’ Bosworth (01:38:13): Amateur. Amateur, amateur.
Amateur. Okay. You also have this website that I came across that I don’t know if people know about it, it’s this funny name. I’m not going to mention that. I don’t know if you want people…
Andrew ‘Boz’ Bosworth (01:38:22): Warden Shortbow, it’s an anagram of my name. Warden Shortbow, an anagram of Andrew Bosworth, wardenshortbow.com. I love photography. I love it. It’s a real passion of mine.
So here’s the question. What’s your favorite photo that you’ve taken?
Andrew ‘Boz’ Bosworth (01:38:33): Art is actually a great place to talk about trusting yourself, by the way. And I know it’s cheesy to say it, but I think Rick Rubin’s recent interviews on what art is and how people make it is spot on. You have to make it for yourself and you have to love it. And if someone else loves it and it finds broader resonance, that’s awesome, but that’s not why you do it. And if you start to try to do it for broader resonance, then you’re kind of chasing something else. It’s media, it’s entertainment, but it’s not art. It’s some other thing. And I say all this to basically tap dance and say, I kind of love a lot of my photos and it’s very hard to pick a favorite. The one that is popping to my head, which has more emotional [inaudible 01:39:12] is a picture I took of my son playing in the street and just jumping in a puddle, wearing a rain boot, rain slick kind of a thing. And it’s not sharp, it’s not in focus. It’s a vignette, it’s an idea.
And I really do think Ansel Adams talked a lot about how the goal of the photographer is to create a capture that expresses to the viewer what it felt like to be there. And people forget that he was a master of the dark room even more so than maybe than the capture, the print. The print was where he did amazing work. And I’ve had the pleasure to go to his dark room in Big Sur and spend time with his son and watch them do development in that room. And he had elaborate scripts of how he would highlight, dodge and burn different parts of the photograph to get it to have the resonance that he wanted. And he fought for photography to be accepted as an artistic medium, which it wasn’t, which I find so resonant in today’s AI art conversations where once again, we’re trying to gate keep what is art and you just don’t get to do that unfortunately.
So this picture of my son, no one would call it a technical marvel, but as a vignette of it capturing for me personally, but also I think in general for parents, the ephemerality of these tremendously touching, charming human moments that you have with your children, that’s the one that comes to mind.
The Two Best Spots in a Career
Lenny: Amazing. We’re going to try to find it and link folks to it. And on the point of Rick Rubin, something he says along the same lines is that you think of art as your diary. I am just describing what I find interesting and important and nobody can come to me and say my diary is wrong. It’s my diary, this is how I see the world, and that’s okay. And that’s where the best art comes from is just emboldening to this… there’s an awesome video of him saying exactly this that I was recently watching actually. Boz, this was so much fun. I am so thankful we made time for this. I’m looking forward to our parenting and relationships podcast in the future. Joking, not joking. Two final questions. Where can folks find you online if they want to follow what you’re up to and how can listeners be useful to you?
Andrew ‘Boz’ Bosworth (01:41:19): Sure. I’m @boztank on Instagram and on Threads and also on X, facebook.com/boz. And I also have my own podcast, which is a technical deep dive, so it’s pretty different, I would say. It’s a technical deep dives to try to go deep on one or two topics each time. That’s called Boz to the Future. You can find that on Spotify or iTunes.
Boz to the Future, buy some Quest stuff.
Andrew ‘Boz’ Bosworth (01:41:44): Get yourself a Quest 3. Let’s be honest. Dude, treat yourself.
There you go. Or these Ray-Ban sunglasses. I’m a big fan. Boz, again, thank you so much for being here.
Andrew ‘Boz’ Bosworth (01:41:52): Cheers. Thanks, brother.
Bye, everyone. Thank you so much for listening. If you found this valuable, you can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app. Also, please consider giving us a rating or leaving a review as that really helps other listeners find the podcast. You can find all past episodes or learn more about the show at lennyspodcast.com. See you in the next episode.
Glossary
| English | 中文 |
|---|---|
| 4-H | 4-H(美国青少年发展组织) |
| Ami Vora | Ami Vora(Facebook/Meta 资深产品经理) |
| anagram | 字母重排(anagram,将原词字母重新排列组成新词) |
| Ansel Adams | Ansel Adams(美国著名风光摄影大师,以黑白风景摄影闻名) |
| Atlas | Atlas(希腊神话中擎天的泰坦神,此处比喻承担重任的人) |
| ATT | ATT(App Tracking Transparency,Apple 的应用追踪透明度隐私政策) |
| Be Kind | ”要友善”(Be Kind) |
| Big Sur | Big Sur(加州海岸著名风景区) |
| binary encoding | 二进制编码(binary encoding) |
| Bootcamp | Bootcamp(Facebook 新员工培训项目) |
| Boz to the Future | Boz to the Future(Andrew Bosworth 的个人播客节目) |
| Brian Chesky | Brian Chesky(Airbnb 联合创始人兼 CEO) |
| Chris Cox | Chris Cox(Facebook 早期核心工程师、高管) |
| Communication is the Job | ”沟通本身就是工作”(Communication is the Job) |
| crest | 纹章(crest,家族徽章) |
| Dave Federman | Dave Federman(Facebook 早期工程师,Boz 的好友) |
| Dave Filoni | Dave Filoni(《曼达洛人》主创之一,Star Wars 资深创作者) |
| Dr. Becky | Dr. Becky(临床心理学家 Becky Kennedy,育儿专家) |
| Eric Schmidt | Eric Schmidt(Google 前 CEO) |
| extreme ownership | 极致责任(extreme ownership,源自美军特种部队的领导力理念) |
| eye of Sauron | 索伦之眼(eye of Sauron) |
| eye relief | 眼距(eye relief,镜片到眼球的距离) |
| Failure Corner | 失败角(Failure Corner,播客中的固定环节) |
| FAIR | FAIR(Fundamental AI Research,Meta 的 AI 基础研究实验室) |
| field of view | 视场角(field of view) |
| Galactic Star Cruiser | 银河星舰(Galactic Star Cruiser,迪士尼星球大战主题沉浸式酒店体验) |
| Gell-Mann Amnesia | Gell-Mann 失忆症(Gell-Mann Amnesia,指对媒体在陌生领域深信不疑、在自身专业领域却深知其谬误的认知偏差) |
| Good Inside | 《Good Inside》(Dr. Becky Kennedy 著,育儿与情绪管理书籍) |
| green belt | 绿带(跆拳道段位) |
| hand tracking | 手部追踪(hand tracking) |
| human in the loop | 人在回路中(human in the loop,人类参与 AI 决策循环的模式) |
| IC | IC(Individual Contributor,个人贡献者,非管理岗员工) |
| identity threat | 身份威胁(identity threat,指自我认知受到质疑时的心理防御反应) |
| J.C.R Licklider | J.C.R Licklider(美国计算机科学家,“人在回路中计算”理念的先驱) |
| Jon Favreau | Jon Favreau(《曼达洛人》主创,著名导演/制片人) |
| JV | 二队(Junior Varsity,校队预备队) |
| Karate Kid | 《龙威小子》(Karate Kid,1984 年经典电影) |
| like button | 点赞按钮(like button) |
| Listening is the Job | ”倾听本身就是工作”(Listening is the Job) |
| LLaMA 2 | LLaMA 2(Meta 发布的开源大语言模型) |
| LLM | LLM(大语言模型,Large Language Model) |
| Lou Holtz | Lou Holtz(美国传奇大学橄榄球教练) |
| Mandalorian | 《曼达洛人》(Star Wars 系列剧集) |
| Mark Sleek | Mark Sleek(Facebook 早期工程师) |
| Matthew Ball | Matthew Ball(知名科技与娱乐行业分析师) |
| mental model | 心智模型(mental model) |
| Meta | Meta(公司名,原 Facebook) |
| Michael Crichton | Michael Crichton(《侏罗纪公园》作者、著名作家) |
| mixed reality | 混合现实(mixed reality) |
| monotype | monotype(等宽编程字体) |
| motion blur | 运动模糊(motion blur) |
| multimodality | 多模态沟通(multimodality,使用多种沟通方式和媒介) |
| News Feed | 动态消息(News Feed) |
| pass-through | 透视(pass-through,头显中看到外部真实画面的技术) |
| Patrick Stewart | Patrick Stewart(英国著名演员,《星际迷航》《X战警》系列等) |
| prom | prom 舞会(美国高中/大学正式舞会) |
| Rick Rubin | Rick Rubin(美国传奇音乐制作人,近年出版关于创意与艺术的著作) |
| RPC | 远程过程调用(Remote Procedure Call) |
| SAV | SAV(系统异常值/问题检测) |
| Scott Trowbridge | Scott Trowbridge(迪士尼幻想工程部高管,负责星球大战主题园区) |
| Sheryl Sandberg | Sheryl Sandberg(Facebook 前 COO) |
| Shire | 夏尔(《魔戒》中霍比特人的家园) |
| State of the Union | 国情咨文(State of the Union,美国总统年度演讲) |
| The Dream Machine | 《The Dream Machine》(M. Mitchell Waldrop 著,关于现代计算机发展史的书籍) |
| Thrift | Thrift(Meta 开源的跨语言 RPC 框架) |
| Tracy Emin | Tracy Emin(英国著名当代艺术家,以霓虹灯文字装置艺术闻名) |
| Veritas | Veritas(拉丁语”真理”,亦为哈佛大学校训) |
| wall posts | 留言板帖子(wall posts) |
| 沃伦·巴菲特 | Warren Buffett 的公认中文译名 |
Reformatted by reformat_english.py
打造 Meta | Andrew ‘Boz’ Bosworth(CTO)
访谈记录
Lenny: 你基本上是 Facebook 的第 10 位工程师。我想那时候一定有很多人们不太听说的艰辛和痛苦。
Andrew ‘Boz’ Bosworth: 我每次睡觉从不超过四个小时。每隔四个小时就会醒来查看报告,看有没有人在攻击网站。电影里可不会告诉你这些。
Lenny: 你每周工作 120 个小时,没有任何爱好。
Andrew ‘Boz’ Bosworth: 我不想抹杀其中的浪漫色彩。只是大多数时候,我们听到的那些浪漫故事都来自成功者。全身心投入一件事、承担风险,这本身是健康的,但在当时绝不光鲜。
Lenny: 动态消息(News Feed)。那是你在 Facebook 早期的项目之一。当时人们并不想要它。显然,他们错了。
Andrew ‘Boz’ Bosworth: 不过 News Feed 其实比人们想象的要容易应对。所有人都在愤怒的同时,他们使用产品的频率立刻翻了一倍。
Lenny: 从经济价值来看,Boz 在 News Feed 和广告上所做的事情,其交集创造了 1 万亿美元的价值。
嘉宾介绍
Lenny: 今天的嘉宾是 Andrew Bosworth,也就是大多数人都知道的 Boz。Boz 是 Meta 的首席技术官。他在 2006 年初加入了当时还叫 Facebook 的公司,是最早的工程师之一。在 Meta 任职的 18 年间,他打造了互联网历史上最具影响力和最重要的产品,包括 Facebook 的 News Feed——这是社交网络中第一个算法排序的内容信息流,基本上就是人们心目中 Facebook 的样子。
他还搭建了 Facebook 最初的移动广告平台,并持续运营了四年。他还参与构建和扩展了 Facebook 的消息系统、个人主页、Timeline、Facebook 群组,甚至内部的工程师训练营。最近,他担任广告和商业平台副总裁,领导工程、产品、研究、分析和设计团队。2017 年,他创建了公司的 AR/VR 部门,即现在的 Reality Labs。
如今,Andrew 负责 Meta 在 AR、VR 和混合现实领域的工作,以及 Quest、Ray-Ban Meta 智能眼镜等消费级硬件产品。
在这次广泛的对话中,我们谈到了很多重要的经验和故事:Facebook 早期到底是什么样的,为什么你应该更多地向上级寻求帮助,为什么沟通本身就是工作。还有 Meta 过去几年转型中的经验教训,Boz 对 Apple Vision Pro 的看法,大量关于领导力和职业发展的建议,打造第一个 News Feed 的经历和从中得到的教训,以及成功和失败的故事,还有很多很多。
如果你喜欢这档播客,别忘了在你喜欢的播客应用或 YouTube 上订阅和关注。这是避免错过未来节目的最好方式,也对播客帮助极大。
那么,广告之后,我为大家请出 Andrew Bosworth,aka Boz。
[赞助商广告已跳过]
开场趣闻
Lenny: Boz,非常感谢你来做客。欢迎来到播客。
Andrew ‘Boz’ Bosworth: 谢谢邀请。我一直是你节目的忠实听众,也一直在关注你发布的所有内容,很高兴终于有机会参与进来。
Lenny: 我也是,能请你来我真的很兴奋。我至少有十亿个问题想问你。不过我想先从几个关于你的趣闻开始,我一个个说出来,然后你挑一个最想讲的,讲讲背后的故事,怎么样?
Andrew ‘Boz’ Bosworth: 好,听起来不错。
Lenny: 好。你参加过 14 场 prom 舞会。
Andrew ‘Boz’ Bosworth: 没错,这是真的。
Lenny: 好,我继续说。好。
Andrew ‘Boz’ Bosworth: 哇,这个开场可够猛的。
Lenny: 这个可能就是你要选的那个了。你大学时是全国跆拳道冠军。你在大学时当过马克·扎克伯格一门人工智能课的助教——不过据我了解,这其实并不是你进入 Facebook 的原因。哈佛大学曾招募你去打橄榄球。你小时候在 4-H 俱乐部非常活跃,养动物、带着它们参加县里的集市展览。你曾和 David Copperfield 同台。
Andrew ‘Boz’ Bosworth: 这是真的。
Lenny: MC Hammer 曾经夸你的穿着很时尚。乔治·W·布什总统夸过你的鞋子和锃亮的脑袋。
Andrew ‘Boz’ Bosworth: 没错,这些都是真的。我想说,哇。首先我得确保大家明白,我是作为绿带获得的全国大学冠军——这就相当于二队冠军。先让大家清楚这一点。
Lenny: 赢了就是赢了。
Andrew ‘Boz’ Bosworth: 重量级实战对打。我跟你说……prom 舞会的故事挺有趣的,它和 4-H 的故事有关。我当年是个铁杆 4-H 会员,入选过全国 4-H 名人堂,在那边做了很多事。
说到这个,4-H 是一个非常棒的青少年项目,男女都参加。我当时跑遍全州、全国各地,参加领导力活动、各种会议,做大量的公开演讲。
而且几乎每一个 4-H 活动最后都有舞会。很多人不知道这一点。会议结束时、夏令营结束时——你真的会去露营一周——到最后总有一场舞会。
prom 舞会高手
因此,想要去很多 prom 舞会,最重要的一件事就是——我得会跳舞。而且事实证明,至少在 90 年代,女生挑选 prom 舞会伴侣的首要标准就是:他愿不愿意跳、会不会跳?而我的答案是肯定的。再加上我认识很多不同学校的女生,这就是成功的秘诀——如果你的目标就是这个的话。高二那年我去了两场,高三去了 12 场。有一次我一个周末连赶三场——周五、周六、周日晚上各一场。
Lenny: 关于你还有一个有趣的事实——你基本上是 Facebook 的第 10 号工程师,加入的时间远远早于它成为明确的成功故事。我想那个早期阶段肯定有很多人们不常听到的痛苦、挣扎和艰辛。大家看了《社交网络》那样的电影,觉得”哦,太有意思了,我也得创个业,说不定也能做成万亿级别的公司”。我很好奇那些早期日子到底是什么样的。有什么特别深刻的记忆吗?
Andrew ‘Boz’ Bosworth: 嗯,关于”第 10 号工程师”这个说法其实有个小笑话——我和另外五个人是同一天入职的,而之前已经有九位工程师了。我们同一天加入,所以大家都是第 10 号工程师。所以具体是第 10 到第 16 号,取决于你怎么算。
我在博客里写过这段经历,也经常讲这个故事:那段时光确实很有趣,建立了深厚的战友情谊和难忘的记忆,但这些都是在极其高压的熔炉中锻造出来的。
那时候,我们几乎所有的人都住在离办公室一英里以内。大部分饭菜都是一起吃的——因为我们在工作。倒不是说我们不是朋友,而是因为工作的缘故,就是——哦,吃完饭直接回去继续干活。
早期创业的真实面貌
有些事情你当时不会意识到,比如——没有人能帮你。没有专家。所以不是那种”嘿,我遇到一个棘手的问题,该找谁聊聊?“答案是——没人。你得自己跟自己对话,把问题想明白。或者”天哪,服务器容量不够了。“那就——“好的,你得去 Fry’s Electronics 买一堆零件,自己组装一台新服务器,然后把它们跑起来。可能还得自己开车去托管机房,上架,再回来继续运维。”
人们真的很低估一件事:今天你即使去一家中等规模的公司上班,尤其是在为创业公司服务的生态系统已经如此繁荣的情况下——薪资、财务、IT、HR,这些事情在很多情况下都有专业人员来处理。
在 2000 年代初,根本不是这样。就你一个人,你的私家车,还有你愿意花在上面的时间。
所以我并不是要否定那段经历的浪漫色彩,只是我们听到的大多是成功者的浪漫故事。我们很少听到那些在二十多岁时真正牺牲了大量社交和娱乐生活的人的故事。
我的付出得到了回报,所以没人替我感到遗憾,也不应该替我遗憾。但还有其他人做了完全一样的事情——也许他们更努力,也许更聪明,也许做得更好——但最终没有成功。这确实是一个巨大的牺牲。
所以,我很理解人们对创业的热情,我喜欢创业文化。我觉得人们愿意全身心投入一件事、承担那种风险,这是健康的。但创业在当时并不光鲜。回过头看,你会觉得”哦,那段经历挺……”我们给它加了一层光环。但在当时,你完全感受不到什么光鲜。
Lenny: 对。在你提到的那篇文章里,你说你每周工作 120 小时。
Andrew ‘Boz’ Bosworth: 对。
Lenny: 没有任何爱好,还胖了很多。
Andrew ‘Boz’ Bosworth: 对。我们还喝了很多酒来弥补,所以胖了很多……而且饮食也不健康。那段时间真的很疯狂。
我之前可能讲过这件事。我最早做的东西之一,是一个反垃圾信息、反爬虫的防御机制。但我们没有任何运维支持,没有 7×24 小时的在线支持。
所以我做了这个工具后,必须每四个小时醒一次。大约两年时间里,我从来没有一次睡超过四个小时。每四个小时就要起来查看报告,看看有没有人在攻击网站。如果有,我就得爬起来反击回去。如果没有,那很好,回去继续睡。但这些东西,电影里是不会告诉你的。
Lenny: 这简直比养新生儿还累。
Andrew ‘Boz’ Bosworth: 而且没人要求我这么做。甚至都没人要求我去做反垃圾信息、反爬虫这件事。我就是觉得这是个问题,然后去做了,这就是我想到的解决方案。如果我是个更优秀的工程师,也许我会想出一个更好的方案。
给创始人的建议
Lenny: 那也许可以作为这个话题的收尾——当你和创始人交流时,你会给他们什么样的建议?
Andrew ‘Boz’ Bosworth: 我想对此保持谨慎,因为我对创始人说的话……我首先会告诉他们的是,我从来没有当过创始人,我需要承认这个区别。我是 2006 年 1 月 9 日加入的,那时 Mark 创办公司已经快两年了。我没有参与过融资,不需要处理那些方面的事情,甚至不需要应对董事会或业务层面的问题。在某种意义上,我很幸运能在那个时候加入。
但我会首先告诉创始人:“我对你的建议,你应该有所保留地听取。我并没有真正站在你的位置上过。”
如果可以称赞你一下的话——你的节目有一个我特别欣赏的地方。我们这个行业有一整套专业人员的体系。我在硅谷成长起来的那些年,大家总在谈论 ACM,也就是美国计算机协会。你听说过那些传奇性的专业组织,它们为我们的领域提供支持。
但由于技术变革的飞速步伐,以及这些机构的参与方式——甚至连学术界、广义上的学术圈——都有些脱节了。你从那些地方获得的工具,在我们的领域并不实用。
所以我确实认为,我们彼此之间提供的指导一直是一个关键的、可持续的资源。到了今天,终于有了像你的播客和Newsletter这样的资源,真正致力于帮助行业内的专业人士——这在过去十五到二十年里几乎是缺失的,我很高兴看到这种变化。因为如果你是一个新人,过去你真的得认识某个内行,才能向他们请教问题。
所以我帮助创始人的时候,很多时候我能帮他们看战略,能帮他们思考技术选型,能帮忙看业务、管理、组织架构。但我也尽量说清楚,有很多东西我确实从来没有接触过。
所以,即便我们刚才在说 2006 年加入 Facebook 的普通工程师有多么不容易——天哪,2004 年 Mark Zuckerberg 的处境大概更加艰难。那个故事肯定已经被人讲过了,但即便如此。我觉得这两者……几乎在所有规模上都一样。无论你把时间线拉回到多早,挑战都很有意思,都值得讨论。
最常给出的建议:善用你的上级
Lenny: 你最受欢迎的帖子之一,是你分享的一段话,关于你经常给出的建议。你说的是:“在我的职业生涯中,作为管理者、董事会成员、顾问和朋友,我发现自己需要给出的最频繁的建议,就是让人们更直接地善用自己的上级。“能聊聊这句话的含义和具体表现吗?
Andrew ‘Boz’ Bosworth: 这其实是一件非常正常、自然、健康的事情。顺便说一句,我们在私人关系中也这么做。就像我在那篇帖子里写的,我们想自己来完成,想亲自做,向所有人证明自己能搞定。
我们心里会想:“如果我求助了,是不是等于放弃了那个目标?是不是等于承认自己在’证明自己能独立完成’这个顶级目标上已经认输了?”
但我们经常忘记的是,你的工作往往不是亲力亲为。你的工作是让事情完成——把事情做完、做好、做对、做得到位。而很多时候,你完成工作所需要的资源,就在你的经理、你的合作伙伴、你的顾问、你的导师那里。这些资源就在他们手上。
作为管理者,有多少次我碰到这样的情况:有人被告知”这是你的任务”,他们说”收到”,然后去做,回来交差,结果做错了。我并不是在怪他们做错了——他们没有跟我确认,理解有偏差,我们之间沟通出了问题。这个锅我来背,没问题。
但结果就是,六个月过去了,事情没有做对,因为他们一开始就没理解任务要求,我们在任务说明上没有沟通清楚。
还有一种情况,他们回来了,但花了一年时间。我就问:“为什么花了一年而不是六个月?“他们说:“哎呀,我有好多事情要处理。“而如果他们当时给我发封邮件,我完全可以帮他们扫清障碍,铺平道路。我可以说:“不不不,这个不用操心,集中精力做那件事。“那我们就能提前六个月完成,他们也不会那么沮丧。所以,关键是要保持轻量级的沟通。
当然,我也认为作为管理者,我们有责任说:“嘿,这本身就是工作的一部分,你得自己想办法搞定。“我一直告诉我的管理者们,我们最有力的一种做法就是拒绝拍板。有人拿一个事情来找我,很多时候我们觉得自己有义务发表意见、帮忙决策。我会说:“不用,我觉得你已经掌握了。我觉得你面临的挑战是对的挑战,你的处理方式也是对的。放手去做就好。“就是这样。所以我们这些做领导者、导师、顾问、董事的人,也有这方面的责任。
不过话说回来,这种事在私人关系中也一样。你和伴侣在一起,你想用正确的方式做一件事,但你没有跟对方沟通。你其实是在冒很大的风险,而回报却很小。你去问一句”嘿,这个你想让我这样做吗?我不太确定”——他们不会生气的。
所以我觉得挺有意思的,我们在脑海里建了那么多空中楼阁,筑起一道道隔墙,阻止我们去利用身边那些本来就是为了帮助我们成功而存在的支持体系。我昨天刚看到一个很棒的说法。Patrick Stewart,我最喜欢的演员之一,他演的角色我也非常喜欢。他谈到演员去试镜的事。这对演员来说是很残酷的,对吧?你去三十个、四十个试镜,被一次次拒绝。很艰难。大家多少都听说过这些。但他说:“没有人比你试镜的那个人更希望你成功,因为他们希望你表现优异。因为一旦你表现优异,他们的工作就完成了。他们巴不得你才华横溢。”
你的经理也一样。没有人比你的经理更希望你表现出色,因为当你做得好,你的经理——他的日子就好过了,她的日子就好过了。所以我觉得我们常常陷入一种心态:“不不不,他们在考验我。“不是的。他们在为你加油。我向你保证。
如何向管理者寻求帮助
Lenny: 我很喜欢这个建议。我想人们之所以不这样做,正如你所说,是不想让经理觉得自己不知道在做什么,或者觉得自己解决不了问题。你有没有什么建议,关于什么时候适合主动去找经理求助?
Andrew ‘Boz’ Bosworth: 对于那些在这方面比较胆怯的人,我觉得可以给沟通加一个框架,让你的经理很容易参与进来。你可以说:“嘿,这件事我在推进中。目前卡在这里,这是当前的进展。“甚至可以说:“如果这些看起来都没问题,你不需要回复。如果有你觉得我应该做得更好、做得不同的地方,或者你觉得能帮上忙的,告诉我。”
我特别喜欢那种文字版的、五到十个句子的邮件——“无需回复,这是目前的状况。“即使一切都很顺利,我也会觉得:这个人理解紧迫性,理解任务要求,还在给我一个心跳信号,一个小小的状态更新。
然后,如果两周后,假设卡住的问题变得严重了,你就可以说:“嘿,抱歉,我现在确实需要你的帮助了,我在这件事上确实被卡住了。“而此时我已经有了上下文,脑海里已经有了你在那边埋头苦干正确事情的模型。即便在你被卡住的时候,你也可以让我的工作变得超级轻松。比如:“嘿,我想请你帮个忙,如果你能给这个人发封邮件,这里有个草稿,大致是这个意思,那就太好了。”
或者,把具体的问题列出来:“我觉得这是你想要的。对吗?是还是不是?如果不是,好的,我们再花时间讨论。如果是,那就没问题了。“这样一来,我不仅了解了最新进展,有了心理模型,而且已经参与进来了。同时,你让我帮你的成本变得极低。大家总是很惊讶——为我工作的人总是很惊讶,当我告诉他们,我工作中很大一部分就是在做这些零碎的小事。
在我这个规模上,有点像是在转盘子。根据不同的统计方式,我手下有一万到一万五千名员工。所以就像时不时我得拿起一个全新的盘子、一根全新的杆子,真正投入精力。但大多数时候,我只是尽量把每件事都碰一下,保持运转的势头。所以如果哪个盘子掉下来了,而没有人告诉我”嘿,我们这里转速在下降,动能在流失”——那我就会很沮丧。“啊,那个盘子掉了,我现在又得在这边重新开始一个。”
所以我觉得大家低估了这一点。他们对我的工作的理解,和我实际的工作……我的工作其实就是无数次轻量的触碰。
Lenny: 所以我觉得这里的一个关键要点是,第一,要更倾向于向你的经理和领导求助。我很喜欢你说的这种表达方式——不一定要说”这是我需要你做的事”,而是”这是目前的进展,这些可能是阻碍我的问题,这些是我的疑问,这些是正在发生的事。”
这其实跟我发现的一个很有效的方法很相似,我简单分享一下——这个想法叫”Lenny 状态”邮件。我每周都会给我的经理发这样一封邮件。“Lenny 状态”,有点像”国情咨文”的感觉。内容就是:“这是我当前的优先事项,这是我大致在想的事情,然后这些是我需要你帮忙解决的阻碍。“
HPM 周报
Boz: 我们以前其实有一种固定格式,叫做 HPM。Highlight、People、Me——重点、人员、自己。从大概 2008 年到 2014 年,Facebook 的每个经理都会发给他们的上级,甚至发给整个领导团队。我记得有一次,我负责我们叫做 comm apps 的业务时,我直接发给了 Zuck 和整个领导团队。
内容就是:重点包括什么?把那些大事列出来,你需要知道的。人员方面,谁遇到困难了?谁有风险?谁做得非常出色需要表扬?然后是自己,你个人状态怎么样?我们就叫它 HPM。
说来也有意思,我已经很久没想起这个了。但没错,我觉得这类做法是有效的。而且你看,每个经理都不一样,所以在 Meta 这个层面上也是如此,这也是另一个成功之处。我觉得大家的倾向是对每个经理都采取同样的方式,但这行不通,因为每个经理都不同。
但你可以问每个经理:“你喜欢怎么接收更新?“你刚开始和他们合作时就可以问。比如:“你的节奏是什么?你喜欢以什么方式了解情况?“对我来说,我会定期做一对一沟通。随着组织越来越大,这些沟通间隔变长了,所以大家改用更多书面形式来替代。
但没有哪个经理会因为你说”你喜欢怎样了解我的工作进展”而生气。这是一个完全合理的提问。
Lenny: 你分享的那个具体做法我特别喜欢——就是帮对方起草一封发给另一个团队负责人的邮件,写好”这是你需要告诉他们的话,这样能真正推动这件事”。这个想法太棒了。
Boz: 顺便说一下,我总是会加上自己的……我不会直接复制粘贴。我会看着那个草稿说:“好,我明白了。“其实很多时候关键不在于你想让对方听到什么,而在于语气和调性。
这里面包含了很多背景信息。比如,你是不是已经跟对方来回沟通了十二个回合了,那这封邮件在他们看来会不会觉得我太过强势?还是说,“嘿,你第一次听到这件事,是我的疏忽。这是我们的情况,需要你的帮助。”
所以,这很大程度上不是为了让我方便复制粘贴。其实,你在语气和措辞中包含了大量信息,能帮我把这件事落好,而不是让人觉得我突然冒出来、压迫感很强。
信息透明与团队信任
Lenny: 这涉及到一个我经常听说的、Meta 工作方式中非常核心的东西,就是透明度。任何人都可以在问答环节向 Zuck 提问。公司鼓励大家不断在内部发布自己的想法和工作进展。所有数据都是共享的。当然这也经常导致泄密,我听说你对此深恶痛绝,这是你的一块心病。
Boz: 就感觉是对团队信任的背叛。就是感觉……我从小踢球运动长大的,美式足球、英式足球、田径。你简直无法想象你的一个队员把战术泄露给对手。你能想象你会怎么做吗?“你被踢出队伍了。对不起,你不能待在这里了。“好了,抱歉,你继续。
Lenny: 对,而且现在人这么多,很难找到是谁泄的密。所以以这个弊端为例——而且我猜还有其他弊端,也需要投入大量精力,还经常让人当场难堪——你认为透明度带来了哪些好处?为什么它是 Meta 工作方式中如此重要的一部分?
Boz: 这其实回到了一个原则:真正优秀、有才华的人,你要充分发挥他们的能力。你要确保他们被充分利用。所以,任何时候当他们掌握了错误的信息,或者根本没有信息,你就阻塞了公司拥有的经济价值最高的东西之一——就是这个人的时间、注意力和才华。不仅如此,你还会让他们更加沮丧,他们离职的可能性也就更大。
如果任何一家公司的命脉是其中的人——这些人共同致力于某个目标和使命、协同工作——那么你就想最大化这种潜能。而打造这种真正开放的信息生态系统,就是我们实现这一目标的方式之一。在我们公司,很多出色的、令人惊叹的成果并非来自自上而下的指令,而是来自人们不仅理解我们自上而下要达成什么,还掌握了远更多的信息来采取行动。
自上而下与自下而上
人们总在谈论自上而下还是自下而上的文化,在我看来这其实有点像神话。如果你见过 Mark Zuckerberg,你就知道这不是一个自下而上的模式。我们追求的想法,首要的是 Mark Zuckerberg 的想法。这并不是说他不接受新的想法,他当然接受,这也是一种自下而上的形式——人们可以把想法带给他,他消化吸收后决定是否采纳。但当他自上而下提出方向时,他不是在微观管理,但他确实深入细节。我对此要谨慎表述。
但他确实为你创造了空间,让你可以把他在说的东西做出三四个版本带回来,然后他再从中塑造和打磨。而你要做到这一点,前提是……如果你没有自由度,那当然不行,但如果你没有信息也不行。否则,如果你不了解我们想达成什么、为什么要做、基础设施怎么样、资源可用性如何、性能表现会怎样——那你就只能按剧本走。对于高才华、高创造力、高投入的人来说,这非常无聊。
当然,这确实要付出巨大的代价。你必须非常善于处理涌入的信息。大多数公司的大多数人,会消费掉给到他们的所有信息,但信息本身是经过精心管控的——他们拿到的是被筛选过的、原本就打算让他们看到的信息。
我们把这套做法完全颠倒了过来,听起来很美好,但并非没有代价。即使是从外部加入公司的高管,我需要辅导他们的一个重要事项就是:如何在所有噪音中找到有价值的信号?
你必须有一套管理信息的系统。你必须有一套对涌入信息进行分诊的机制,把那些在错误渠道的、跟你不相关的大量信息过滤掉,搞清楚哪些群组你想加入但只在你选择时才去浏览——
Boz: ……哪些群组你想加入但只在你主动选择时才去浏览。以及哪些东西你是通过推送通知接收的?那些是实时的事项,这需要花一些时间来适应。
Lenny: 你关于自下而上和自上而下的观点非常有趣,因为当我想到 Meta,我印象中它是一个非常自下而上的文化。我听说每个人都能提出自己的想法,做实验,被鼓励去尝试各种东西。所以听到你说大部分重大想法其实来自上层,真的很有意思。而且细想一下也完全说得通。
Boz: 这属于一种迷思。我不认为它完全错误……作为一种契约,它确实比许多其他公司更自下而上,因为你在这个框架内确实拥有一定的自由度。但别搞错了,像 Mark、我、Chris Cox 或 Javi,我们对公司应该做什么有着非常强烈的意见,你的自下而上是在那个框架内运作的。不过有一点是真的:你可以问 Mark 任何问题,他都会回答你。我也是,Chris 也是,Javi 也是。而且我们确实会从与员工的讨论中获得启发。所以我觉得,事情并不像人们倾向于描绘的那样非黑即白。
Lenny: 我认为这里最大的启示之一是,至少要让人们感觉自己有很多发言权,即使公司实际上非常”这就是大的战略方向”,你们非常擅长让人们觉得自己可以产生影响、拥有话语权。
清晰的边界与护栏
Boz: 我跟你说,最重要的一点就是给人们清晰的指导方针,让他们知道自己的空间在哪里。比如,“哪里有发挥余地,哪里没有?“我们在跟 Mark 的评审中,或者我的团队跟我的评审中,有一点我很明确——对于 UI 的某个部分,我会说:“这部分我来画给你看,就要做成这样。“而对于另一个部分,我会说:“嗯,那个也很重要,但我没有清晰的构想,你来发挥吧。“所以就是设立非常清晰的护栏——“好,哪些地方我们只是指派任务?哪些地方有更多灵活性?”
Lenny: 你能想到一个具体的例子吗?比如你深入到细节、亲自画屏幕的情况?
Boz: 多年来有不少这样的例子。早期我们做动态消息(News Feed)的时候,Mark 在白板上亲自画了前端团队需要实现的每一个像素。但在后端,他就说:“做排序,搞个排序系统就行了。“所以我觉得自己那时候挺幸运的,我就说:“太好了,我去搞排序的东西。“而 Chris Cox 和其他人则得一个像素一个像素地去对。不过也不总是这样。快进到现在,我们讨论排序的时候,Mark 并不是一直放手不管的。当我们开始现代化我们的排序系统——这在过去五年里一直在做——Mark 深度参与其中,他会问:“各项权重怎么分配?你是怎么衡量不同因素的?”
所以它是双向的。就我个人而言,我深度参与了一些相对冷门的领域。比如,我非常坚持要把手势追踪和混合现实做到头显里。这么说吧,团队里并没有太多支持者。当然我们有手势追踪团队——他们非常出色——也有混合现实团队,但有很多人觉得这些功能对于这个设备走向大众市场并不是关键。但我一直坚信,从易用性的角度来看它们是关键。所以我强硬地推动这件事,没给任何人回旋余地,并对手势追踪的性能基准设定了非常高的标准。团队跟我说不可能。结果并非如此。它做得非常好。
创始人与高管该深入到什么程度
Lenny: 这涉及到这个播客中经常出现的一个话题。就是关于创始人和高管到底应该深入到什么程度的争论——是该放权、赋能,还是”不,我们就这样做,我要亲自过问每一个设计稿”。而且总会有这种反复:一开始说”好,让大家放手去干”,然后事情往往开始不太顺利,于是”我们要收回控制权”。您对什么时候该深入有什么看法?创始人和高管应该怎么思考这个问题?
Boz: 对创始人来说,这个回答可能没什么用——取决于具体是什么细节。有些细节真的非常重要,有些则真的没那么重要。我得说,这不意味着它们完全不重要,你必须去做它们,只是它们不是决定成败的关键。确实有我很尊敬的人,Brian Chesky 上过节目说过:“Airbnb 只会做我能亲自参与的工作,做到什么程度取决于我能做到什么程度。“这是一种非常极端的做法。我很尊敬他,也很欣赏他们的运作方式。我认为,如果你有非常优秀、值得信任的人才可以负责更大的模块,那是一种选择。如果有一些机制可以让你有效地检查进度、确保方向正确,那是另一种结构方式。
而且 Airbnb 肯定也有一些 Brian 不亲自管理的工作在做——财务、会计、人力资源。所以显然存在一些我们信任他人去处理的非技术领域。或者法务。这些领域我们是相信它们在正常运转的。所以从我和创始人们的交流来看,很多人后悔放权太多。我完全理解。或者说他们把某个关键的东西放权出去了,结果那恰恰是最重要的东西。对我来说,判断标准就是:“最重要的是,你如何确定什么东西最关键?“
索伦之眼
在 Meta 内部,我们开玩笑地称之为”索伦之眼”(eye of Sauron),到现在还是这么叫。当 Mark 认定你正在做的东西是最重要的事情时,没有哪个细节小到他不会注意到。他会在一场评审中,同一场评审里说:“从战略上看,我们的方向偏了。另外,这个像素肯定是错的,你必须修掉。“跨度非常大。坦白说,有点自夸的话,我为自己也能做到同样的事情感到自豪。和我共事的人经常评论说我们的领导风格——我觉得 Chris Cox 也是一样——就是会在真正重要的事情上深入到极低的层面。而其他很多事情当然也重要,我们很高兴它们在做,但要么它们有相当清晰的产品路线图,要么行业里有明确的参照,要么那就是一个你必须有但不会决定成败的功能。所以先把它做到差不多,然后再迭代,就可以了。所以我认为,到底要深入到什么程度,确实取决于具体的细节是什么。
Lenny: 太有趣了,我用过完全一样的比喻——索伦之眼——在谈论 Airbnb 里 Brian 非常看重的事情的时候。我给人们的建议是:“你在职业生涯中不想在索伦之眼下面待太久,因为如果你一直做最重要的东西,你会精疲力竭。但你也不想离得太远。你不想待在夏尔,但你想待在附近。”
Boz: 没错。两者都有代价。我从2012年到2017年做了好几年广告业务,管理那个大型广告和商业平台团队。这个领域当然非常重要,但当时 Mark 有太多其他事情要处理——移动端转型等等——他确实把它委托给了我。那段经历太棒了,能获得他那样的信任真的非常酷。但同时,你也时刻提心吊胆,因为”Mark 不了解这些东西。万一这一切都是……”我的下属们会担心,因为他们很久没有跟 Mark 做过评审了。就是说,你在索伦之眼的凝视强度下会痛苦,但在它缺席的阴影中同样会痛苦。没有完美的位置。
Lenny: 太搞笑了。我在想中土世界里哪个地方可以比喻那种状态。
动态消息的诞生与争议
Lenny: 好的,你刚才提到了动态消息(News Feed),那是你在 Facebook 最早的项目之一。关于动态消息,我知道几个有趣的事实。第一,它是所有社交网络中第一个算法驱动的动态消息,可能也是同类产品中的第一个。第二,它是 Facebook 内部编写的第一段 AI 代码,用于对动态消息进行排序。所以当时创下了很多个”第一”,而动态消息显然后来成为了一件大事。如今人们想到 Facebook 时,本质上想到的就是动态消息,但它刚推出时争议极大。人们非常反对这个功能,他们不想和别人分享这么多信息——至少他们自己是这么认为的——然后他们最终发现,“哦,这其实正是我想要的。“你在经历那种建造出人们最初拒绝、后来才意识到其实这正是他们想要的东西的过程中,学到了什么?
Boz: 这个故事其实在你访谈中经常被提起,核心就是”你必须对自己正在构建的东西有信念。“你在选择你的用户,就像你的用户在选择你一样——我有时会这样想这个问题。而你确实会看到一些初创公司犯的错误是:他们获得了一批早期用户,但这批用户的需求实际上把他们引向了一个与更大市场正交的方向。于是他们就被自己的早期用户绑架了。我们一次又一次地对产品应该是什么样子有一个愿景,而那个愿景并不是我们当时正在交付的东西。所以那些正在使用我们当时产品的人并不确定那个变化是不是他们想要的。但我们有信心,随着时间的推移他们会接受的。我们不总是对的,但在这些情况下我们是对的。
现在回顾起来,动态消息其实比人们想象的要容易应对,因为每个人都在愤怒的同时,立刻就把产品使用量翻了一倍。所以我们有一个优势——所有人都在说”我太讨厌这个东西了”,然后不停地刷新、刷新、刷新。我们当时就想,好吧,等等,这里存在认知失调——他们表述的偏好和经济学意义上的显示偏好之间存在差距。所以对我们来说,坚持动态消息比外界想象的要容易一些。但人们有时会误解这件事。他们觉得”哦,教训就是不要听用户的。“完全不是这样。我们非常在乎用户的反馈。甚至在动态消息这件事上,我们也确实搞砸了一些东西。我总是开这个玩笑——就像你在派对上,音乐很响,你跟人说话,然后音乐恰好在你音量最大的时候停了,于是全派对的人都听到了你说的最后那句话。
你本来是在一个公共场所说的,所以那也不算什么私密的话,但你也没打算用那个音量向所有人广播。我们基本上对整个用户群做了这件事。我们把之前发在留言板(wall posts)上的内容——当然,本来任何人都可以去那个个人主页看到——然后把它放到了主信息流里,像现在年轻人说的”放到主页上”(Put it on Main),然后有人就说”啊!“所以确实,我不想把这说成……我们确实搞砸了一些东西。这不是说”哦,这只是执行上的小瑕疵。“还有一个需要判断的事情是:你什么时候只是犯了一个小错误,什么时候犯了大错?什么时候是方向本身错了,什么时候只是细节错了?这是一门艺术。一门真正的艺术,而且你不总是能像我们这样有用户数据来判断。
所以很大程度上取决于”你是否对预期发生的事情有一个清晰的愿景和直觉?“然后实际发生了什么,你能否诊断两者之间的差距?在动态消息的案例中,我们犯了一堆小错误,但方向本身是对的,我真的为我们当时做的工作感到骄傲。我和 Chris Cox 在工程层面可能是最核心的两个人,你在产品经理(PM)那端也看到了——在此之前没有任何排序过的信息流。我之前确实为反垃圾信息等方面建过一些 AI,但相当初级。不过那大概是同类网站中第一个面向消费者的内容 AI。我们打造了搜索之外人类历史上最高效的变现界面,我认为是这样的。对于那些好奇的人说明一下,我用变现这个词不是因为我觉得钱是最重要的东西,但我确实认为它体现了你创造的经济价值,而这与人类效用之间有很强的正相关性。当然,我尊重有些人可能不同意这个观点。
Lenny: 就经济效用而言,Boz 主导的动态消息和广告的交集,创造了上万亿美元的价值。
Boz: 这不是——
Lenny: 干得漂亮。
Boz: 也不能说少了,确实不少。
Lenny: [听不清]
Boz: 我们为这份工作感到骄傲。
热情与职业探索
Lenny: 你在一篇关于动态消息的文章中有这样一段话:“它比生命中任何事情都更彻底地占据了我。它让我认识到一个真相:当你对某件事充满热情时,你会做出更好的工作、更聪明的工作,效率会高出一个数量级。”
Boz: 热情是无法替代的。在那篇文章之后,我对自己又有了新的认识——就是我发现自己在本质上是一个容易对事物产生热情的人。这是一种天赋,我很幸运拥有它,我也理解不是每个人都这样。实际上广告业务就是一个好例子。当 Mark 让我去做广告的时候,我说”不,我不想做。我觉得我对那东西没有热情。“我对自己有一个固定的认知,我有非常强烈的自我认同——我就是这个做 AI 基础设施和产品的人,我在这个领域耕耘,不想改变。但我错了。我就是一个会对事物兴奋起来的人。一旦深入广告领域,我就觉得”哦,这太有趣了。这是一个三方市场,有各种不同的……”感觉就像在跟行业里其他玩家下国际象棋一样,我为此超级兴奋。
然后当他又让我去做硬件的时候,我说”不,我不行。我是个软件人。Mark,我是个搞软件的。“而现在,我热爱这份工作。那是一个如此迷人的领域,我学到了很多。所以我觉得那句话是对的——当我找到自己充满热情的东西时,效果确实很好。那之后我学到的是,给自己一些空间去理解自己能否对某个事物产生热情。当然,我之前的工作中也有些部分我始终没能找到热情,六个月之后我就不得不离开了。真的是这样——要么辞职,要么被开除,我做不好工作,对工作也不在乎。所以我确实有这种自知之明。并不是说我对任何事情都能产生热情,但事实证明我的兴趣面其实相当广泛。
Lenny: 我觉得这是一个很有价值的职业教训——“不要假设你不会对某个可能出现的事物产生兴趣。“在这方面你有没有什么想和大家分享的?比如”去探索一下,给自己六个月时间,看看你能否对它产生热情?”
Boz: 完全同意。我的职业轨迹在某些方面其实很不寻常——很长一段时间里,我几乎每六个月就换一次工作。我在做动态消息(News Feed)相关的诚信工作。然后我全职做动态消息(News Feed)大约一年。之后做网站速度优化和基础设施,检测 SAV 和各种问题。然后做 Bootcamp,再做消息和群组。我有一个很有意思的感受,我开玩笑说——对于年纪够大、记得《龙威小子》的人来说——我觉得自己就像在不停地刷栅栏、给车打蜡,但最后我学会了空手道。到最后,我终于得到了回报——因为我一路走来,认识了很多人,在不同的领域工作过,理解了各种不同的运作机制。而和我同一批加入的人,他们都在升职,但都走的是单一轨道。他们就待在一个地方,然后升职。而我一直在到处转。在职业生涯早期,大概有段时间看起来我的进度比同龄人慢。然后当我终于拐过那个弯——真的就是从接手广告业务开始,我做了五年——我的职业一下子就垂直上升了。简直像坐火箭一样。从那以后我就一直保持这个势头。所以我最常给人的建议是,至少对我而言,我从中学到的教训就是——我愿意拼命地学。我之所以调动,是因为我觉得学不到足够多的东西了。我无聊了,学不到新东西了。而最终到了广告那个岗位,以及我现在做的这个岗位,最棒的地方在于——这些工作让我在五年里学到了海量的东西,从来没有停止过学习。你偶尔会遇到那种特别深的工作,可以一直学下去。
与此同时,我很多朋友,他们的职业早期轨迹比我更好,但后来他们真的对自己做的事感到厌倦了,却无处可跳。没有别的去处……他们成了某个领域的专家,但那个领域对他们自己来说已经被榨干了。也许他们甚至比自己想要的多待了一阵,因为待着很舒服,或者因为公司需要他们。结果到了职业生涯中期,这反而成了他们的障碍。所以对我来说,建议就是——“跳进新事物里,给自己六个月。如果那不是你的菜,没关系,你积累了大量新技能,我保证这些技能迟早派上用场。继续前进。“同样,在职业生涯早期做这种跳转时,要以学习为优先,以学习为优化目标。想想复利。复利的前十年看起来没那么惊艳,但十年之后就开始好看起来了。
Lenny: 我很喜欢这个建议。我自己常给的建议也很类似——“经历的多样性往往最终成为你长期积累中最有价值的东西。“就是去尝试各种不同的事情,做一些内部工具,也许做做客户支持,信任与安全,面向用户的产品,基础设施——我从产品经理的角度想,也可能是工程师或其他职能。顺着这个思路有一个问题。我们之前聊到了索伦之眼(eye of Sauron)和做公司最重要的事情。关于职业生涯中应该花多少时间待在那个中心位置,你有什么建议吗?
职业中最好的两个位置
Boz: 嗯,在其他条件相同的情况下,我认为有两个特别好的位置。第一个是,在公司没怎么关注但你很重要的领域,承担大量的重量级工作。而且量一定要大。你得真正把那些事情扛下来,在那边搬山填海,因为我告诉你,作为管理者,当有一座大坝在挡着洪水的时候,你会对那个撑住大坝的人充满敬意。你会说:“你继续撑着,Atlas,那边干得好。“第二个最好的位置,或者说同样好的位置,就是做最重要的事。做最重要的事的时候,就要用到 Eric Schmidt 给 Sheryl Sandberg 的那个建议——“这是一艘火箭,先上去再说。别问自己坐在什么位置,上去就行。”
如果是最重要的事,你分到的一块可能会比较小。每个人都想去。拿到你能拿到的那块就好。如果是最重要的事,拿到你能碾压、能做好的那块,从中成长,因为你会获得极高的曝光度,获得大量经验。你会看到烈火中是什么样子——真刀真枪——那是无价的。你在任何地方都用得上这些经验。所以我会这么说。这是在项目选择阶段的建议,但我接下来要提醒一句。你要理解,那些从烈火中开始的项目,希望最终能锻造成型,冷却下来,不再在火里——但愿如此。同样,那些挡着洪水的大坝也有开裂、崩溃、被洪水冲垮的倾向。所以我认为,在选择阶段,你要争取处于这两个位置之一,但之后……你要坚持走下去。
再说一遍我的观点——如果你没有投入感,如果你做不出出色的工作,如果你不热爱它,那就离开。如果你已经把它耗尽了——曾经热爱但现在不再热爱了,那就离开。如果你仍然热爱、仍然投入,那很好。这是一件好事。你完全可以从锻造炉走到大坝,再走回来,随时间推移来回轮转。你不必总是不停地跳到最新的火场上去。我在广告业务之后试过这么干一次。实际上,我花了六个月时间,我们在 2012 年打造了第一个移动广告产品,基本上拯救了那次 IPO——当时情况已经变得相当严峻了——
Lenny: 对,我记得那个时期。
Boz: ……到了那个节点。我跟 Mark 说:“这太有意思了。也许你可以一直让我这么干,每六个月把我放到最大的火上。“他转过头来对我说:“Boz,那不是一份真正的工作。”
他说:“我需要你留在这里,把这个事情推进下去。“我照做了,之后的四年半都待在那里。那段经历非常棒,非常棒。说真的,我对他的……说好笑的,我这样说会挨批的——我既是我们公司最大的批评者之一,也是他最大的粉丝之一。这两个身份我都有。但今天我们聊的这些,我觉得 Mark 确实展现得非常好。他在我的职业发展中,把我推到了各种我自认为无法胜任的地方,他看到了其中的机会,让它成为了现实。
如何给 Mark 反馈
Lenny: 关于这一点你学到了什么?给 Mark 提负面批评的时候,有什么是他会接受的吗?你有什么心得?
Boz: Mark 对所有的信息和所有观点都有着极其强烈的渴求。有一点很有意思,我之前提到过,作为创始人,我认为你必须拥有巨大的信念,尤其是创始人。你必须拥有极度的自信。我认为 Mark 是我见过的意志力最强的人,纯粹从意志力的角度来说。关于 Mark 有一件事很有意思——你给他反馈,他会听。他是一个很好共事的人,你给他反馈,他会认真听。真的。但大多数时候他会告诉你你错了,以及为什么错了。这就是最常见的情况。然后接下来会发生的是……这真的很神奇。就像在接下来的一两周里,你会看到一些变化。我不觉得他是刻意在这么做……
Boz: 我一直开玩笑说,信息总是会传到他那里。每天有大量的信息涌向他。然后到了晚上,他会用所有这些信息重新编译整个世界,第二天再回来。顺便说一句,这不仅仅适用于产品方面的工作。在我脑海中,我想的是产品相关的事,比如我会说”嘿,我觉得这个产品在这件事上做错了”,他会说”不不不,这就是为什么它不是那样的”。然后产品就开始发生变化。给他关于他自己开会时的表现或表达方式的反馈也是一样,他会说”哦,我当时那样做是因为……”,然后解释原因。再过几周,你遇到类似的情况,他会调整自己的出场方式。所以我其实觉得给他反馈是一件很令人满意的事。真的很有效,但你确实需要着眼于长远。他会……他做的每件事都不是偶然的。他做每件事都有他自己的理由。
Lenny: 这很好地诠释了”观点坚定,执念灵活”(strong opinions, loosely held)。
Boz: 对,没错。
Lenny: 这也让我想到,你刚才用了编译器的类比,我在想这更像模型训练。他一夜之间重新训练了自己的模型。
Boz: 对,没错。有趣的是,Mark 有一个特点特别好笑——如果你早上给他一些反馈,接下来他开的六个会,不管是不是关于那个产品的,他都会问别人对那个反馈怎么看。他不会说是你提的,他就像”嘿,你们觉得这个产品里的这个东西怎么样?“所以你会在和他开会的时候看到他这么做。他来找你开会讨论别的话题,突然会冒出一句”嘿 Boz,你觉得这个产品、这个想法怎么样?“就这样,他会在一天之中拿着那个小想法不断去压力测试,而且他特别爱三角验证——去寻找你可能没看到的各种不同视角。所以他真的非常重视每个讨论议题的多元视角,这一点很有意思。
Lenny: 在为他的模型获取更多训练数据。我懂了。
Boz: 你别想让我管 Mark 叫大语言模型(LLM)。这不公平。[听不清]
Lenny: 我们都希望能像 Mark 一样聪明。你说话的时候我注意到了你的纹身,这让我想起你至少有两个我所知道的纹身。
Boz: 哦,对。
Lenny: 一个是加利福尼亚的轮廓,我完全理解。加州是一个非常特别的地方,但你还有另一个纹身,就两个字——Veritas。你能聊聊这是什么意思,为什么对你重要吗?
Veritas 纹身与对真实的执念
Boz: 关于纹身这件事,好笑的是我从高中毕业的时候,我……我不知道我当时算什么类型的人,但我到 21 岁才第一次喝酒。我以前是一个非常循规蹈矩的人。我会想”为什么要去纹身,改变自己的身体?""为什么要染发?顺其自然就好。“这里有一部分原因,我想我是个挺幸运的人,对自己是谁、想成为什么人有很大的自信,这没问题。但在某些方面,我也以一种现在回想起来很不可思议的方式去评判别人,今天肯定不会那样了。但当时来说,去纹身对我来说是一件很大的事。然后我突然意识到,身体不过是承载我生命的载体,你可以随心所欲地对待它。它不是什么……它是你所拥有的东西,如果你想装饰它,就可以装饰它。
所以去纹身对我来说确实是一件大事,我完全转变了对自己身体的看法,对别人的身体、他们的存在方式乃至时间的看法,某种程度上甚至是对死亡的理解。就像”嘿,生不带来死不带去。一切终将逝去……”当你 18 岁的时候,你觉得自己会永生;等你到了 22 岁,作为一个饱经沧桑的 22 岁”老兵”,你会想”唉,赶紧纹吧,反正都是要完的。“所以是的,这就是我为什么纹了 Veritas,拉丁语中”真理”的意思。我得说这有一点俗气,因为这也是哈佛的校训。但我用的是 monotype 字体。就是程序员用的那种字体。
纹身对我来说还有一件有意思的事,它也是代际变迁的一部分。我们成长的那个年代,纹身在成年人眼里是和帮派、摩托车手、水手或某些特定人群联系在一起的。现在据我了解,我最近看到一个统计数据,我这代人中有纹身的人已经超过了没有纹身的人。所以我觉得我们在文化上也发生了转变……我觉得丰富的自我表达方式很美好。我真心这样认为。所以我完全支持这一切。
Lenny: 从你的描述来看,我的理解是,“真理”这个概念对你的思维和工作方式非常重要。
Boz: 在这一点上我的名声确实在我前面,恐怕是这样。我觉得当我年轻的时候,我把诚实看作……顺便说一下我是错的。我把诚实看作一种免死金牌——只要你诚实,你想说什么就可以说什么。事实根本不是这样。我之前写过这方面的内容,我最大的职业遗憾,远超其他,都是因为我不够友善。我以前认为……我之前写过一篇叫”要友善”(Be Kind)的笔记。表现得”nice”(讨人喜欢)——那是居高临下,是跟人说半真半假的话,是敷衍了事。我反对那样。但”kind”(友善)不是那样的。友善是”嘿,我怎样以一种真正有建设性、有帮助的方式传达这个反馈?一种能帮助对方而不是仅仅让他们感到沮丧和无助的方式?”
我觉得在年轻的时候,我在这件事上做错了很多。所以诚实仍然是我性格中很大的一部分。认识我的人绝不会说我不诚实。我觉得大家理解并尊重我的直率,如果我有担忧或问题,我会提出来。只是我现在更善于提出这些问题,并且表达出真正的关心和信念——如果我不认为我们可以做得更好,如果我不认为我们可以解决问题,如果我不相信这个局面,我就不会提出来。所以诚实仍然是我身份认同中很重要的一部分,这也是我很引以为豪的。但我必须说,我表达诚实的方式自从纹下这个纹身以来已经发生了巨大的变化。
“沟通本身就是工作”
Lenny: 这听起来很合理。这其实触及到一个我非常想聊的话题,也是你最经典的文章之一,我最初就是通过这篇文章认识你的——那篇叫”沟通本身就是工作”(Communication is the Job)的文章。
Boz: 对。
Lenny: 我知道很多人读过这篇文章,也有很多人没有。我很想请你聊聊这意味着什么,为什么重要,以及你为什么想分享它。
Boz: 好的。这是那些……尤其如果你有志于成为领导者的话……领导力不是管理,领导力也不是成为唯一的责任承担者,甚至不总是和问责制同一回事。但如果你想产生影响……
Boz: ……它甚至不总是和问责制同一回事。但如果你想对你周围的世界产生影响,唯一的方式就是通过创造能够影响他人的作品或表达来实现。仅此而已。如果你想产生影响,如果你想创造某种持久的改变,这就是全部途径。它可以是你的小小的人际关系中,可以是你的团队中,可以是你的公司中,也可以是整个世界中。归根结底都是沟通。你常常听到人们说,“哦对,我一年前就写过这个了。“就好像,“是,但你一年前写得不够好,否则我们不会白白浪费一年没有去做这件事。”
人们总觉得,“哦,我有那个想法,“就觉得这有什么意义了。它毫无意义。它绝对没有任何意义。或者有人说,“哦,我写了这篇帖子。“但你没有让它真正传达出去。这是你的问题,不是受众的问题。人们总想责怪受众。但受众就在那里。我之前甚至提过这一点,希望大家注意到了,当我说,我给某人一项任务,他们六个月后回来,做的东西是错的,我会认栽。我会自己承担这个责任。这对他们来说也不好,他们会因为浪费了时间而沮丧,但正如我说的,这是我的责任。我没有清楚地沟通我想要什么,期望是什么。
他们自己能帮自己吗?当然可以,这需要各方共同努力。我们应该从两个角度一起来解决这个问题。我还有另一篇帖子叫”倾听本身就是工作”(Listening is the Job),就是这件事的另一面。但我真心相信”沟通本身就是工作”(Communication is the Job)……实际上,它与从美国海军陆战队和海豹突击队中发展出来的”极致责任”(extreme ownership)理念有关。就是每当出了问题,我会问自己,“在沟通方面,我本可以做什么不同的事来让结果更好?“我是否可以更好地设定优先级?我是否可以更好地设定期望?我是否需要给团队一个更好的指标?我是否安排错了人?
组织架构图即沟通工具
顺便说一下,我常说的是,组织架构图就是沟通工具。它们并不真实存在。你和你经理之间并没有一根物理上的线。它们只是沟通工具,用来让人们大致了解事物的组织方式以及该去找谁。所以所有这些都是沟通。沉默也是沟通,我没有主动联系你关心你、关心你的项目,这也是沟通。我们之前谈到过那种信任,那意味着什么?那意味着信任。那意味着责任。不主动检查本身就有含义。你不可能不沟通。你总是在通过你的表情、你的穿着、你的身体语言在传递某种信息。你在传递什么?
我给你一个有趣的例子,希望我们能把它放在播客里,因为如果你看的是视频版,你会注意到我的摄像头一直在不停地调整光线。不是太暗就是太亮。我在试用一个新摄像头。我是个技术宅,我尝试很多摄像设备,尝试很多麦克风设备。我喜欢各种最新的小玩意儿,所以我在试新东西。但效果不好。我脑子里就在想,这在传递关于我的什么信息?人们会觉得我不在乎,或者觉得我不够称职。
理解他人的心智模型
这就是我说的意思。所以现在,我感到有必要在播客中解释一下,这样我就能清楚地传达事实并非如此。我在职业生活中所努力做的,很大程度上就是去理解他人的心智模型。他们现在在哪里?我指的是具体的人,比如我的经理或我的关键技术负责人,我也指的是广义上的人,比如团队,我还指的是更广泛的普通人。在这场对话中,他们处于什么位置?我如何塑造我的语言、我的形象、我的人设、一切,来引导他们从他们所在的位置走向我想把他们带到的地方?
这需要我对想把他们带到哪里有非常清晰的想法,也需要对他们现在在哪里有清晰的想法。而且我想告诉你,这没有听起来那么费劲。我想没有人会说我有一个精心伪装的人设。不是那样的。而是要对人们所处的起点抱有极大的共情。这是对我而言最重要的事。其他的,比如我在会议中如何表现、努力多微笑,因为我是个高大让人害怕的家伙,这些都是你可以慢慢练习的小事,它们会变成第二本能,很容易做到。真正困难的是对受众抱有共情——“他们在哪里?他们从哪里出发?“而当你没有做到时,要为此承担责任,承担极度的责任。
Lenny: 这里面有太多好的建议了。有太多线索我想跟进,但我们就顺着最后这条——试着理解对某个人最好的沟通方式是什么。能不能举个例子,让它对大家来说更具体一些,比如你做了什么,“哦,原来我要这样和这个人沟通”?
Boz: 我给你举几个例子。一个是多模态沟通。有句老话说,重复不会让祈祷失效。我认为大多数有经验的沟通者,无论是写作者还是公众演讲者,都会谈到多次、以多种不同方式重申一个观点的重要性,以确保人们有机会内化它。你要直接说出来,你要用比喻。所以对我来说,我会先做一个全员大会,然后写一篇包含全员大会内容的帖子,因为不同的人对这些不同的模态会有不同的反应,在不同模态上吸收信息的速度也不同。这是一个比较简单的例子。
直面恐惧与疑虑
另一个我一直在想的事情,是确保你回应了人们的恐惧和担忧。如果人们觉得你不知道发生了什么,他们是不会听你的。所以,我最喜欢做的事情之一,当我们讨论某个问题时,我会一上来就说,“让我先说清楚,这是我们正在遇到的问题。我知道我们遇到了这个问题。我知道什么是最重要的。“然后我会说同样的话——但如果没有前面那段,他们实际上会完全忽略我,因为如果他们不接受前提,怎么可能信任你的结论?你懂我的意思吧?我觉得这里面有很多值得说的。显然,当你们面对面时,这要容易得多,因为你可以阅读面部表情。即使在现在这个视频通话中,我看到你在点头,我就知道,“好的,他跟上我了。“然后我会加一句,“你懂我的意思吧?“但如果你给我一个歪头的表情,我就会再举第二个例子来把观点说透。但你要一步步积累。大多数人在职业生涯中,现实地讲,会从试图影响一两个人开始。那就是你的起点。一两个人。那就是你需要沟通的对象。你的经理,一个队友,这就是你拥有的。然后你逐步积累、积累、再积累技能,在越来越大的规模上做到这一点。
Lenny: 我太喜欢这些建议了。我觉得这对人际关系也很有帮助。“你 upset 的是这件事。”
Boz: 完全同意,百分之百。
Lenny: “这是我认为我们可以做的。”
Boz: 再说一次,我在 Meta 获得了非常优渥的支持来做自我提升——有很棒的导师,Sheryl Sandberg、Mark Zuckerberg,还有很多人以及教练——这些绝对让我成为了一个更好的伴侣和丈夫,我妻子也是。而且反过来说也一样。有了孩子之后,我深入阅读了大量育儿方面的文献。顺便恭喜你,对了。深入研究那些文献让我成为了一个更好的管理者。在思考人们如何管理自己的情绪,以及如何在那些时刻与他们互动方面,绝对让我成为了一个更好的管理者。
Lenny: 太棒了。我们需要出第二版 Boz 的育儿建议。
Boz: 没错。
Lenny: 还有感情建议。
Boz: 全是好东西。《No Bad Kids》、Lansbury、《Good Inside》、Dr. Becky。都是——
Lenny: 我喜欢 Dr. Becky。
Boz: 我真的认为现代育儿经典作品非常丰富。
Lenny: 太棒了,这么多好内容。
Vision Pro 与 VR 头显
Lenny: 好。你提到了那个小设备和那些你喜欢玩的相机。我们来聊聊 Vision Pro 和 VR 头显吧。你试过 Vision Pro 吗?怎么看?
Boz: 实际上,Mark 和我是一起试的。首先我想说的是,当这款头显出来的时候,我们松了一口气,因为它里面的东西并不代表什么根本性的突破。它内部的一切都是我们大概率也能做出来的,除了 Apple Silicon——那确实是个奇迹,但值得指出的是,Apple Silicon 在处理缩放显示分辨率这类任务时虽然有大约两倍的优势,但不幸的是分辨率的性能需求是二次方增长的,所以两倍的线性缩放优势并没有你想象的那么大。这是第一步。但我们仍然假设,以那个价格,加上他们传奇般的对细节和打磨的执着,他们大概率做出了一个很棒的产品。我一周前在自己的播客上跟 Matthew Ball 说过一句话——“听着,我已经准备好上市后说我们有性价比最高的头显。如果你想要一款在价格范围内最出色的头显,我们有,就是 Quest 3。”
但当我跟 Mark 并排试用 Apple Vision Pro 之后,我真是太兴奋了,我们觉得,“不,我们认为我们实际上拥有最好的头显。“当然,我们不是说它在所有方面都是最好的。如果你坐着不动看一部高分辨率电影,Apple Vision Pro 确实非常出色。非常出色。分辨率的优势体现出来了,透视(pass-through)的调校在你保持静止、直视前方时也非常亮眼。他们在 UI 方面也做了一些非常漂亮的东西。不过这恰恰也是让我们有点恼火的地方——作为产品人,我们确实会有些恼火。
我知道这种情况我们都会遇到。Apple 会遇到,Google 会遇到,我们也会。我们内部有一堆一直在把玩的东西,将来某天发布的时候,会被指控说偷了别人的创意,而实际上我们并没有。如果你感兴趣,可以去看看我在 Quora 上关于点赞按钮(like button)历史的回答,之前就发生过类似的事情——我们在别处上线之前就已经在内部构建了点赞按钮。总之,这是另一回事了。这种情况在我们这个行业经常发生,我确实不应该这么在意。这多少是我自己的自尊心在作祟,如果我负责任的话,应该把它控制住。
那些漂亮的 UI 打磨,他们在眼动追踪方面做得非常出色。关于眼动追踪有一个有趣之处——要做到他们那种方式,这就是为什么你必须使用处方镜片插入件(prescription insert),所以它不支持戴眼镜。你必须专门配处方镜片插入件,而且很贵。他们可以让追踪眼睛的摄像头透过镜片以及镜片周围的光线进行拍摄。而我们的 Quest Pro 是从侧面进行追踪的,这样你就可以佩戴矫正镜片。
像这样的不同选择各有利弊,但仍然很酷。他们能把这个做进去确实很棒。同时,我们的手部追踪(hand tracking)更好。当然,应用库我们本来就知道会更好——这对他们不太公平,他们才刚发布,出货量还很小。但我就是觉得舒适度……最让我注意的是,Apple Vision Pro 的视场角(field of view)真的很小。网上有些人对它的描述是不准确的,他们是在离镜片很近的位置做的测量。一旦你把眼距(eye relief)——也就是镜片到眼球之间的距离——算进去,对于几乎所有脸型来说,它的视场角相对于我们的产品就变得相当窄了,我觉得这很让人分心。
他们的显示屏比我们的暗得多,而且在混合现实(mixed reality)使用场景中,运动模糊(motion blur)让我非常分心。正如我早些时候在播客中提到的,我是一个超级混合现实爱好者,我对这个方向的潜力非常热衷,原因其实跟他们完全一样——我认为手部交互和混合现实让这种技术对更多人来说更加平易近人。不过我还是很高兴我们的产品线中有手柄,因为它确实扩展了你能做的事情。我们不会只用一种方式操作电脑——我们有键盘、有鼠标,我们一直都在使用多种输入方式。舒适度、我们透视中没有残影和运动模糊、显示屏的亮度——我觉得,“天哪,如果你让我选一个带走,我会选 Quest 3。”
当然,有人理直气壮地说,“这太偏颇了。“当然偏颇。去形成你自己的判断吧。但让我受不了的是,大多数人根本没做过这个对比。他们没有试过 Quest 3。这才是最让我受不了的。如果你去试试 Quest 3,问问自己,你是想要七台 Quest 3——一台给自己、六台给你最好的朋友——还是一台 Apple Vision Pro。我确定不是每个人都会选 Quest 3。有些人确实有适合他们生活的使用场景,Apple Vision Pro 更适合他们,我没意见。
Quest 3 的功能被低估
Andrew ‘Boz’ Bosworth: 但人们甚至不知道 Quest 3 可以做远程桌面(Remote Desktop)。你可以通过一个叫 Remote Desktop 的应用来实现,这个应用非常受欢迎;你也可以进入 workrooms,让你有三块环绕你的显示器,从你的电脑直接推流过来。我觉得有些情况就是,人们根本没下过功夫去了解。他们甚至都没试过。我欢迎所有觉得我有偏见的人,自己去验证你的判断,但前提是你得戴上 Quest 3 好好体验一番,因为它真的是一台相当棒的设备。而且你还省下了 3000 美元可以做很多事。
顺便说一句,他们是怎么做到的?我们发布了一款 1299 美元的头显,人们就炸了锅。他们那款 3500 美元,倒好——“没问题,挺好的,没人在乎。“我不知道该怎么解释。要求公平是奢望太多,我也不在乎这个。Apple 赢得了他们建立的伟大品牌,这是实打实的。我认为那非常了不起。我当然欣赏 Apple 的很多产品。我是他们作品的超级粉丝,是他们所做一切的超级粉丝。这也可能是我对 AVP 期望更高的原因。
Ray-Ban 智能眼镜
Lenny: 好了,我给你看看我最喜欢的 AR 设备,就是这副 Ray-Ban 眼镜。我真的自己买了一副。
Andrew ‘Boz’ Bosworth: 是的。
Lenny: 来,我戴上。接下来我要做一件事——录制。我事先没告诉你我要这么做。我们一边聊,我一边录制。你看这个。
Andrew ‘Boz’ Bosworth: 我喜欢这个。你戴起来也特别好看,很配你。
Lenny: 我岳母说,“你看起来好成熟好有品味。戴上这个真的很有学问的样子。“我们当初买这副其实是为了拍孩子,而不是举着手机挡住 Cameron 的脸。效果真的很棒。
Andrew ‘Boz’ Bosworth: 确实是最好的方式。看着手机屏幕的同时,真正的现实被夹在中间,这真的很难受。不,透过镜片才是正确的方式。
Lenny: 你看,我的新造型。我打算以后一直戴着眼镜了。
Andrew ‘Boz’ Bosworth: 我们得给你升级多模态功能(multimodal)。我从去年十二月就在玩这个功能了——你可以用摄像头让 Meta AI 助手分析你看到的东西,真的很好用。我最近和家人在一个滑雪小镇,我戴着眼镜,就说了一句,“嘿,看看,告诉我你看到了什么。“我发现了一个指示牌,它就给了我指引:“洗手间在右边下那层楼梯。你想吃东西吗?在右边。“它分辨不出我在哪个小镇,但能说”你在某个滑雪小镇,这里是各项设施的位置。“我当时就想,“哇。“这里面有一种真正的魔力。
Lenny: 我觉得我播客访谈也需要这个,这样就能有个小声音告诉我该问什么问题、讲到哪了。
Andrew ‘Boz’ Bosworth: 对,完全没错。
Lenny: 那该多有用啊?
全功能 AR 眼镜
Andrew ‘Boz’ Bosworth: 好了,这个问题我要稍微”作弊”一下说——显然我们已经聊了有一阵了——但体验具备完整 AR 功能的眼镜,我们有一款据传很快就会在内部推出的产品——传闻如此之盛,甚至可以说基本已经确认了。真正有趣的是,能把这些实际上就是”时间机器”的设备拿在手里把玩。技术令人惊叹。有人在做全公司级别的大型演讲时,所有笔记都显示在眼镜上,而且只需一个手势就能控制幻灯片。
Lenny: 天哪。
Andrew ‘Boz’ Bosworth: 令人兴奋的事情正在发生。未来看起来相当光明。
Mark 回应 Apple Vision Pro 的视频
Lenny: 还有一件事我想聊聊。Mark 发布了那个关于”我对 Apple Vision Pro 与 Quest 的看法”的完整视频之后,很多人就说,“天哪,他发这个视频,一定是非常害怕目前的局势,这个做法不对。“当时的考虑是什么?是有战略考量,还是他只是觉得,“这就是我的看法”?
Andrew ‘Boz’ Bosworth: 这个人对现代互联网生态真是了解。大家都戴着 Meta 滤镜看一切,觉得什么都是四维象棋。那真的就是 Mark 对那个产品的真实看法,就是他怎么想的。我觉得他想确保人们记住,嘿,Quest 3 确实是更好的设备,而人们甚至都没有试过。
所以我们这边并不是总在玩四维象棋。有时候我们就是,“这是一件我相信是真实的事,我要用嘴说出来。“他就是这么做的。没那么复杂。当我这么做的时候,所有人都觉得理所当然,因为我的公众形象或个人品牌就是如此。我猜当 CEO 这么做时,人们反而觉得意外。好吧,我理解。那也挺好的。社会对不同角色有不同的期待,这我懂。
我们都熟悉 Apple 在《纽约时报》上登”欢迎 IBM”的广告,还有 Slack 对 Microsoft 做了同样的事,以及 Ballmer 关于 iPhone 的那些评论。这些都不是对产品技术优劣的真正讨论,全都是大张旗鼓、凝聚士气的行为。但 Mark 那次不是这样的。Mark 对此有非常深的理解。他是这方面的专家,我也是。我对我们的选择非常有信心。
顺便说一下,当我使用 AVP 的时候,我能完全代入那个设计者的思维。通过使用体验,我能告诉你那个团队被下达了什么指令,他们在优化什么,他们受到了什么约束。从他们做出的所有选择,我能推断出所有这些信息,而且我理解这些选择在逻辑上是一致的。我们做出了不同的选择,这不应该让任何人惊讶。我们更认同自己的选择。我们本来也可以做他们那样的选择,但我们没有。我们做了这些其他的选择。
所以对我来说,那个重量、那根总是在蹭我耳朵的线、那个可以揣进口袋的电池——我理解。但我不会那么做,而且我知道这一点,因为我有机会那么做,但我选择不这么做。我不知道。我不知道为什么大家会感到惊讶。这不是什么精明的战略举措。就是 Mark 有机会试用了一下,然后觉得,“天哪,我觉得我们应该告诉大家这里真正的故事是什么。“然后我们就这么做了。
Lenny: 我很喜欢这个洞察。我知道很多看到的人都会想,“哦,天哪,他说得对。我之前没这么想过。“所以我觉得这产生了很多那样的影响。
Meta 的低谷与反弹
我想稍微拉远一点,聊聊 Meta 过去几年的历程。感觉上,公众对 Meta 的评价和股价都经历了一个巨大的低谷,然后在过去两年里又有了一个巨大的反弹。这些时期总有太多值得学习的东西。举个例子说说股价——我刚查了一下,跌到了大概 80 美元左右,而今天是 487 美元。
Andrew ‘Boz’ Bosworth: 是的。
Lenny: 我很好奇,从经历那个低谷到反弹的过程中你学到了什么。我知道这个过程还在继续,但就是想问问,你从这段历程中学到了什么?
Andrew ‘Boz’ Bosworth: 是的。嗯,可以学到的东西很多,我告诉你。我们经历了历史上最大的单日股价暴跌,18 个月后又迎来了股市历史上最大的单日涨幅。
Lenny: 天哪。
Andrew ‘Boz’ Bosworth: 正如传奇教练 Lou Holtz 所说:“你赢的时候绝没有他们说的那么好,你输的时候也绝没有他们说的那么差。“Mark 一直在内部引用这句话来指导我们,试图让我们稍微隔绝外界舆论的反复无常。这不仅仅适用于股价,也适用于媒体和新闻报道,在很多方面都是如此。
我跟团队说的一件事——我现在还不得不反复对他们说——就是:“身处其中时很难记住的一点是:你比那些批评者知道得更多。你比市场上的分析师知道得更多。你比媒体知道得更多。你比播客主播知道得更多。你比 Twitter 上的人知道得更多。关于我们公司真正有价值的东西,你比他们都了解得更多。但这并不意味着要忽视他们,因为他们有不同的视角,你需要去理解。即使他们了解的总体不如你多,其中也可能包含你所不知道的部分。“我非常推崇这一点——我会阅读所有的批评,而且读得非常仔细,警惕确认偏差,警惕那些我可能倾向于抗拒但也许是真实的东西。我欢迎一切批评,但我也不会盲目接受。我不会简单地说:“对,这显然是对的。“
Gell-Mann 失忆症
Gell-Mann 失忆症(Gell-Mann Amnesia)是一个值得每个人理解的重要概念。它是指这样一种现象:你读到一篇报纸文章——就说报纸吧,有何不可——关于某个你是专家的话题,然后你会大惑不解,因为这篇文章不仅是错的,连因果关系都是颠倒的。Michael Crichton——我借用他对此的妙语——“这是一个湿路面导致下雨的故事。“你会觉得:“多么糟糕、荒谬的报道。”
然后你翻过报纸的另一页,是另一篇关于你一无所知的话题的文章,你会把它当作福音真理来读。你会想:“哦,看看这条关于那个国外局势的信息。“正如我所说,完全信以为真。我们应该比这更聪明。那这是否意味着你不去读那些东西?不是,你要读。你只是带着这样的视角去读一切:“等等,这是另一种观点。我如何将它整合到一个全面而知情的世界观中?“
长远眼光与沟通
第一点,宏观层面来说,要有长远眼光,意识到当你跌入谷底时,情况没有你想象的那么糟;当你身处巅峰时,情况也没有你想象的那么好。真相始终介于两者之间。第二点是沟通本身就是工作。我认为我们在向市场传达未来投资方面确实做得不够有效。听我说,我们有两个长达十年的重大投资领域。一个是 AI,一个是 Reality Labs。如今 AI 看起来相当不错。我想大家都会认同——有了 LLaMA 2,有了 FAIR,我们取得的那些突破。
人们没有注意到,FAIR——我们的 AI 研究实验室——是 AI 领域被引用次数第二多的研究实验室,仅次于 Google。我们一直在做这些工作。我们不是随随便便就走到这里的。我们一直在深耕,所以现在看起来相当不错。我觉得我们没有向人们充分解释清楚这些。之前核心业务增长强劲,人们愿意忽略它们。用沃伦·巴菲特那句老话说:“只有退潮的时候,你才知道谁没穿泳裤。” 所以当潮水退去——乌克兰战争爆发、利率上升、经济衰退来临——所有人都在为那额外的一块钱拼命,他们会说:“把这些东西砍掉吧。”
我们不得不告诉公司:“你不会想在一个这样的公司工作——时局艰难时就砍掉所有未来增长,只守着核心业务。“这样的公司不过是让自己在某个时刻走向死亡,比预期稍晚一点,但终究是死亡。你想在一个拥有平衡投资组合的公司工作,而我们恰好有。我们没有很好地解释这一点,所以我们花时间向市场、向媒体、向所有人解释了这件事。
现在,我认为随着人们了解了它的规模和范围——当然核心业务也克服了来自 ATT 及其他方面的挑战,这也有帮助——情况看起来相当不错。我确实认为,作为内部人士,第一点是要真正调节自己对外部叙事和情绪波动的依附,这非常重要。而你要基于对自己专业能力的认知来做到这一点。第二点是理解为什么会有差距。差距在哪里?抓住它。通常答案是沟通。
组织扁平化与效率
Lenny: 还有一次大规模的组织扁平化。这是很多人讨论过的事情,管理者变成了 IC。关于如何调整组织以提高效率,你还有什么更多的心得吗?
Andrew ‘Boz’ Bosworth: 当然,我本应该把这部分也放在第一部分里。我有点急于把它漂亮地归结为两点就结束了。但你说得对。
Lenny: 谢谢。
Andrew ‘Boz’ Bosworth: 我们在运营业务的方式上做了重大调整,这非常痛苦。听我说,这要追溯到 COVID 的繁荣时期,当时看起来电商、远程办公以及相关的工具软件——而这恰恰是我们构建的东西——似乎出现了真正持久的结构性转变,这让我们处在了有利位置。所以我们建立了一支庞大的员工队伍来追逐这些机会。我们仍然相信这些机会,但它们回到了原来的时间线上。实际上,如果你看我们内部的一堆图表,就是 COVID 的暴涨,然后——不算真正的暴跌——而是随着它消退,一切都回到了原来的轨迹上。
我们没有失去阵地或浪费时间,但那种”提前拉动”并没有发生。这意味着你的经济学模型不再合理了。你做了一系列投资,但回报在太远的未来,而且加快速度也无济于事,同时你背负着大量额外成本。这很糟糕,真的糟糕。我们对此感觉并不好,真的不好。这就是商业,很残酷,但它会发生。
代际经验的缺失
我确实认为那个时期一个有趣的事情是,对于我们这些亲眼见证了 .com 泡沫的人来说——我在硅谷出生长大,所以当我高中毕业上大学的时候,这一切就在我身边——然后是 2008 年住房危机引发的严重衰退和市场崩溃等等。现在想象一下,你 2009 年大学毕业并找到了一份工作。那么好,你已经工作了 15 年,你可能已经是总监级别了,而你从未经历过一次经济下行。
我认为,除了非常遗憾的——导致我们过度招聘、又不得不纠正的——常规性商业预测失误之外,我们的员工队伍也完全没有”这种事可能发生”的心理准备。这在当时感觉像是天灾,而实际上这是所有企业的周期性规律——这种事在某个时刻总会发生,你希望它不发生、你希望它没有发生,但你必须应对它。所以我认为整个行业在那段时间经历了一场艰难的风暴,而且我觉得我们仍在感受它的余波。我们仍在感受它。
行业不确定性中的共情
Andrew ‘Boz’ Bosworth: 当然,我们很高兴 Meta 已经走过了那个阶段,我们再次增长,以稳定的节奏推进工作,对此感觉很好。但行业中相当多的其他公司就不是这样了。对于工程师、产品经理、设计师以及他们周围所有职能岗位的人来说,这是一个非常不确定的时期。我对此深表同情。我认为,Meta 内部出现的预测失误显然在各处都发生了,而现在,尤其是在利率上升、现金不再那么廉价的情况下,跑道变得更紧了。人们只是在做出务实的判断。我认为我们会从中恢复过来。这是行业正常会经历的事情,但这并不减少我对那些受影响的人、或生活在这种恐惧中的人的同情和共情。
Lenny: 我和一位在 Meta 工作的朋友聊天,我问她 Meta 的工作体验如何,她就说,“很紧张,以前更轻松一些。以前这里那里有些摸鱼的人,“而现在她说,“不,那些人现在都走了。只剩下拼命工作的人,我们工作非常努力。“这让你有什么想法吗?
Andrew ‘Boz’ Bosworth: 是的。我不想评论那些离开的人。人们因为各种不同的原因离开 Meta,同样,岗位裁撤在很多情况下发生是因为我们决定暂时不做这项工作了——我们打算两年后再做,不需要维持一个团队来做。我觉得很难一概而论,因为每一个离开的都是具体的人,有具体的生活和丰富的故事,值得被讲述。但我确实认为,如果有人在摸鱼,而你作为管理者必须做出艰难的决定——你要偏向谁——我深深的怀疑是,算了,我再说一遍,我不认识你说的那位朋友——我深深的怀疑是那个人本身就已经在努力工作了。你懂我的意思吗?那个人本身可能就已经在努力工作了。我不认为我们改变了任何个人的工作努力程度。我真的不这么认为。
我确实认为存在一种选择偏差,影响了谁会留下来,我想你朋友看到的可能就是这个。事实上,如果非要找出一个规律的话,大概就是这样。
失败角(Failure Corner)
Lenny: 也许作为最后一个问题,我有一个环节叫”失败角”(Failure Corner),我请嘉宾分享职业生涯中的一次失败以及从中学到了什么。你有什么想到的吗?
Andrew ‘Boz’ Bosworth: 我失败过无数次。我做过没人用的产品,搭建过无法扩展的技术架构,各种失败都经历过。大部分失败我并不后悔。几乎每一次我都从中学到了东西。要么是通往更好方案路上的一个停靠站,要么是认识到这条路永远走不通——这本身就是一种馈赠。我真正后悔的、认真对待的失败,都是那些个人层面的失败——我以一种自己不引以为豪的方式影响了某个人,也许当时就不引以为豪,因为我没有控制好自己的情绪或心情。我当时感到恐惧,感到害怕,有那么一两个案例特别突出,但我不太方便分享,因为当事人可能不希望我公开。
我分享一个……我觉得我和当事人现在关系很好。我们有过一次非常愚蠢的争论。我记得非常清楚。在早期的客户端-服务器架构时代——Facebook 显然是一个网站,所以你向服务器发请求获取网页,但那个服务器又会去调用其他服务器来收集数据——这是远程过程调用(RPC)的主要使用场景之一,因为动态消息排序是在另一组服务器上完成的,那组服务器有自己的特殊要求、特殊的构建方式和组装方式。所以你的主 Web 服务器会通过远程过程调用向远程服务器发出请求,然后得到响应。我们当时有一个非常破烂的 RPC 系统,我不说是谁写的,反正它就是个垃圾,经常出故障,完全不健壮。我们最优秀的工程师之一 Mark Sleek 构建了一个新的 RPC 框架,叫 Thrift。那是一个非常好的、真正优秀的 RPC 基础设施。
一次关于编码的技术争论
我最好的朋友之一 Dave Federman,我的一个非常好的朋友,也是个出色的工程师,我们在讨论怎么做编码,我说,“我要用二进制编码(binary encoding)。“我说,“你应该用二进制编码。我希望它在传输线上超级高效,“因为这些 RPC 对我来说承担两个任务,一个是活跃的 RPC 调用,但我还把 RPC 存到日志中重放它们,来完成我们在动态消息排序中做的工作。当年就是这么做的,全部是异步离线处理,所以我希望数据尽可能紧凑。我的内存带宽非常有限,而且当时内存非常昂贵。Dave Federman 说,“不不不,Boz,那是短视的想法,我们应该使用 Linux 风格的描述符,用纯英文文本。这样你可以查看日志,能看到内容是什么。内存带宽可能会变得更便宜,但这些日志对开发的可读性才是更好的选择。“这是技术圈的经典引战话题。
在座的工程师们,做过工程师的人都知道,这就像 Vim 和 Emacs 之争一样。这是技术圈的引战话题。这东西能挖得很深。这是老生常谈了。
Lenny: 我太喜欢了,继续说。
Andrew ‘Boz’ Bosworth: 公司一个房间里全是工程师,公司当时也不大,大概公司一半的工程师都在这个房间里。我在大喊大叫,真的,我涨红了脸,满头大汗。我对 Dave Federman 反对我的提议愤怒不已,而我可是 RPC 最大的客户。先说几件事。这事儿太蠢了。Mark Sleek 写两个编码器就行了,这并不难。你选你想要的编码器就行了。这是最简单的解决方案。第二件事,Dave 是对的,顺便说一句。不到一年,内存带宽就完全不再是问题了,相比之下,那些日志完全无法读懂才是真正的麻烦。我不得不额外构建大量自定义软件来解析日志、搞清楚到底发生了什么。而且这件事的核心问题在于,我的身份认同绑在了”我是对的”这件事上。有些人可能不了解,身份威胁(identity threat)是最大的……当你觉得自己处于身份威胁之下时,你最糟糕的行为总是会暴露出来。
当你觉得自我的某个核心部分受到质疑时,你会倾尽全力去捍卫那个自我认知,因为重新定义”你是谁”的代价太高了,所以你会拼命防卫。我的身份认同就是”我是对的”。这导致了我对最好的朋友之一、我认识的最优秀的工程师之一、一个我 literally 住在一起的人,产生了让我至今尴尬的冲突,旁边的每个人都在挠头想,“Boz 怎么了?“我看起来像一个失去理智的疯子。我记得那个房间,我记得我站在房间的什么位置,我记得那个时刻的一切。然后我回家后不得不问自己,“这他妈是怎么回事?发生了什么?“我在问自己,“刚才到底怎么了?“那就是漫长旅程中的众多节点之一——重新理解什么才是重要的,不是要当那个”对的”人,而是要保持开放心态和好奇心,学会如何参与这种竞争。
我当时 22 岁。我现在不会为此找借口,但我记得还有其他几个类似的例子,没有那么技术性,更多是个人层面的,我不会分享,但我对每一个都记忆犹新。对我来说,这些才是真正的失败。
Lenny: 我很喜欢这个故事——虽然已经是很久以前的事了,但它至今仍然留在你心里,产生了这么大的影响。
Andrew ‘Boz’ Bosworth: 天哪,我永远不会忘记那一幕。太丢人了。那句老话说得真好,我知道它是老生常谈,但有时候老生常谈之所以成为老生常谈,就是因为它们确实有道理——“人们不会记住你说了什么,只会记住你让他们有什么感受。“所有人记住的就是你让他们有什么感受。我觉得在那个房间里,我可能让大家感到不安全。那很糟糕。
Lenny: 我喜欢”身份威胁”这个概念。不如就把这期播客命名为”身份威胁”吧。
Andrew ‘Boz’ Bosworth: 说得对。
Lenny: Boz,我们这期节目一开始就准备了无数个问题,现在又多了一个。真希望能继续聊下去,但我知道得收尾了。在我们进入非常精彩的快问快答环节之前,还有什么想和听众分享的吗?
真正的突破:好奇心
Andrew ‘Boz’ Bosworth: 既然我们可能真的会把这期命名为”身份威胁”,那我就分享一下真正的收获吧。我在教练辅导中取得了很多突破——学会识别自己身体在感受到身份威胁时的反应,学习降低其发生概率的技巧和策略,学会在发生时如何应对,以及事后如何修复,这些我都经历过。但我想说,我学到的最重要的一课,是多年之后从观察一个人身上得来的。Ami Vora,一位传奇人物,她很早就加入公司,先是做我们的开发者平台,后来跟我在广告业务合作了很久,再后来担任 WhatsApp 的 PM 负责人很长时间。她离开公司后又做了更多了不起的事。
跟她共事就像在看一个外星人一样,因为她的方式和我截然不同。她可以和一个人有最深刻的分歧——对方说出她认为不仅错误、而且荒谬至极的话——她的回应却是,“太有趣了。你得告诉我你为什么这么想。“我形容不出那种感觉,她是发自内心地这么说。她看到自己和对方之间的鸿沟——可能是个人层面的,可能是专业层面的,可能什么都涉及——她看到他们看待世界的方式和自己看待世界的方式之间的裂痕。她没有把这当成威胁来反应,也没有因为担心自己可能错了、之前可能做错了而感到恐惧,而是以最真诚、最深刻的好奇心去回应。我就这样亲眼看着好奇心把观点之间的高墙彻底拆毁。人们立刻就能感受到她发自内心的真诚好奇,于是也会敞开心扉,变得更加开放。
如果她是对的——顺便说一句,她通常都是对的——对方就会离开时想,“好吧,我在这件事上错了。“但她也会改变自己的想法,这才是关键所在。从那以后,我一直在努力效仿她。当我内心产生强烈的紧绷感时,我会试着拥抱好奇心,“哇,我们看待这件事的方式完全不同。太有趣了。告诉我你为什么那么看。“这可能来自个人反馈。比如有人说,“老板,我觉得你说得太少了。“我会说,“你觉得我说得太少了?真是出人意料。我很想听听更多,因为从来没有人这么跟我说过。“所以我想把这个分享给那些可能在自身行为中看到影子的人。你能做的事情很多,而且你应该去做那些功课。
自我提升的努力总是有收获的、令人满足的,而且会在你生活的方方面面得到回报——与家人、与朋友、以及职业上,我们之前讨论过。而这个关于好奇心的教训是我觉得真正了不起的。在面对挑战时拥抱好奇心,彻底改变了我的生活,我欠 Ami Vora 一份感谢。
Lenny: 哇,我很喜欢这个例子。这本质上就是”是的,然后呢?“的变体,但以一种更……没人会真的说”是的,然后呢?“这是一种特别好的表达方式——“太有趣了。告诉我更多。”
Andrew ‘Boz’ Bosworth: 太有趣了。你一定得告诉我为什么,我想理解它。
Lenny: 我太喜欢这个了。也许新标题就用”太有趣了”。
Andrew ‘Boz’ Bosworth: 就这么定了。
快问快答
Lenny: 好了,说到这里,Boz,我们进入了非常精彩的快问快答环节。准备好了吗?
Andrew ‘Boz’ Bosworth: 准备好了。
Lenny: 第一个问题:你有哪两三本最常推荐给别人的书?
Andrew ‘Boz’ Bosworth: 第一本是《The Dream Machine》,这本书是对现代计算机的史前史做了一次极其精彩的历史梳理,表面上是围绕 J.C.R Licklider 的生平展开的,但实际上远不止于此。在我看来,它填补了历史上一块重要的空白。我觉得在我的专业领域——计算机科学——我们的教育是有缺失的。我们学了艾伦·图灵,学了一些计算机科学理论的技术基础,但现代计算机以及通往它的道路,是一段深刻而迷人的历程,而且在今天格外引人共鸣,因为 J.C.R Licklider 当年的洞见就是”人在回路中的计算”,而我认为我们现在正处于”人在回路中的 AI”时代,两者之间有着极大的共鸣。第二本是《Good Inside》,Becky 医生的书。同样,我认为它是一本出色的育儿书,但不仅如此,它确实包含了关于如何认识和管理自身情绪的方法,这些在任何场景下都很有用。
Lenny: 太棒的推荐。她还有一个线上社区,供……
Andrew ‘Boz’ Bosworth: 她有一个很棒的线上社区,她自己也非常积极参与其中。对于在座的父母们,Lenny,你现在还有点早,但你会到那一步的。我在 Instagram 上读到的那些小脚本,当我在与孩子们面对巨大的情感挑战的瞬间,手边就有现成的话可以说。它们就在我脑海中随时待命。已经提前储备好了,这是一个很大的改变。
Lenny: 这会是我们以后育儿专题的内容。
Andrew ‘Boz’ Bosworth: 没错。
Lenny: 有没有最近特别喜欢的电影或电视剧?
Andrew ‘Boz’ Bosworth: 有的。这个答案很常规——《曼达洛人》。其中一个有趣的原因还是因为我的孩子们,我们和孩子们一起看的,所以有机会去体验了”银河星舰”项目。大家都知道,我是一个彻头彻尾、货真价实的书呆子。我非常欣赏 Scott Trowbridge 在迪士尼星球大战主题园区的工作,也喜欢这个节目本身。在那之前,我们让孩子们——当时九岁和六岁——和我们一起看了所有的电影,然后我们一家人一起看《曼达洛人》,特别有趣。有这种传说体系的故事作为连接纽带很有意思。不仅仅是经典的儿童电影,而是有更深层的东西,而且我觉得对他们来说,他们会觉得自己参与到了一种成人的对话中。我真的很享受这种体验。我也非常欣赏 Dave Filoni、Jon Favreau 以及正在构建那个宇宙的整个团队。
Lenny: 此乃正道。
Andrew ‘Boz’ Bosworth: 此乃正道。
Lenny: 你在面试候选人时,有没有最喜欢问的面试问题?
最喜欢的面试问题
Andrew ‘Boz’ Bosworth: 我每次都会问候选人的一个重要问题是:跟他们共事过的人会怎么说他们的优点和缺点。我喜欢这个问题有几个原因,其中一个是我确实经常去做背景调查,我想交叉验证他们对别人如何评价自己是否有清晰的认知,以及他们如何回应批评。有时候候选人的回答会让我意外。他们会说,“嘿,你会听到很多人这样批评我,但我觉得这并不公平。“这也可以是一个说得过去的回答,但他们得有足够强的底气。或者他们会说,“嘿,这是我正在改进的地方,这是我正在做的事。“但我也很想知道他们认为自己的超能力是什么。面试中人们往往过多关注缺点,我也关心,因为我想知道下限在哪。但对我来说更重要的是,“你擅长什么?什么是那种我只要挂上你的车就能一路飞驰的东西,这才是我想知道的。你碾压一切的超能力在哪里?”
有趣的是,人们很少能准确预测我的背景调查结果。他们对批评意见往往不太准确,但对自己的优势通常说得很准。
Lenny: 有意思。我也非常推崇关注优势,不用太在意自己的弱点。
Andrew ‘Boz’ Bosworth: 但要求人们通过团队来定位自己的表现,这一点非常重要。我们很难靠个人取得多大成就。
Lenny: “沟通本身就是工作”。太棒了。你最近有没有发现什么特别喜欢的产品?Meta 的产品也可以算。
最喜欢的产品
Andrew ‘Boz’ Bosworth: Ray-Ban Meta 眼镜会是一个很有诱惑力的答案。它的多模态沟通(multimodality)功能——我这样说可能要惹麻烦了,因为我在社交媒体上预告了好几个月,目前还在封闭测试阶段,而且”我们在慢慢扩大测试范围”,但”人们已经非常迫切地想要它了”。这大概是最近我体验过的最神奇的东西之一。这不太公平,因为你也确实可以买到 Ray-Ban Meta 眼镜,而且很快——我不说具体时间,但很快——它们就会具备这个功能。不过跳出框框想更有意思。好吧,这个回答有点……我会因为这个回答惹麻烦,因为它有点装。我先声明,我不是车迷。我不喜欢车,也不在乎车。我想要的是一辆能开、不坏的车。我从小坐的车,里程都超过十万英里,因为我们家只有二手车。车总是抛锚——开着开着转向助力就没了,你得拼命打方向盘;刹车会失灵,或者连杆断裂。这些我都经历过。我只想要一辆能开的车。后来我开了一辆本田雅阁,开了十年,然后开了一辆特斯拉 Model S,又开了十年。那辆特斯拉 Model S 停着的时候出了点事,这个回头再说。我买了辆新车,我又想,“我要买辆电动车。“然后我就想,“我要买辆好的,买辆好的,我要买辆好的。“我现在开的是一辆奔驰 AMG EQS,我以前不知道车可以这么好。我从小开的是二手车之类的,我根本不知道这是可能的。我认为这是目前市场上你能买到的最好的增强现实产品。它的抬头显示系统会在三维空间中标注你的转弯。车上有对着你眼睛的摄像头,所以转弯指示在显示屏上的位置会根据你的眼睛位置精确调整。当你快到路口时,就像面前有一面小墙,你得在撞上墙之前转弯。我当时就想,“天哪,我觉得这辆车的增强现实做得太好了。“我不是车迷,也不是在炫耀这辆车什么的。我确实很喜欢它,我不知道车可以好到这种程度,但让我印象最深的是他们在这辆车上把增强现实做得非常出色。
Lenny: 哇。奔驰——增强现实和混合现实(mixed reality)领域的玩家。
Andrew ‘Boz’ Bosworth: 必须的。他们已经杀进来了,而且处于领先地位。
Lenny: 我有时候想搞一个活动,把这个环节里嘉宾提到的产品送给听众,你现在把我的预算完全打爆了。
Andrew ‘Boz’ Bosworth: Ray-Ban Meta 眼镜,你还送得起。
Lenny: 这个推荐太棒了,我得去亲身体验一下。我看奔驰的销量要涨了。下一个问题。
人生座右铭
Lenny: 你有没有一个最喜欢的人生座右铭,经常回想、跟朋友家人分享、觉得很有用的?
Andrew ‘Boz’ Bosworth: 有。说来挺有意思的。我们家的座右铭——就是我、我妻子和孩子们——就是”相信自己”。就是相信自己。其实这个座右铭的由来是这样的:我们家有几件不错的艺术品,其中一件是 Tracy Emin 的作品,她是一位著名的英国艺术家,专门做霓虹灯作品,那件作品上写着”Trust yourself”(相信自己)。它就挂在我们卧室走廊里,我和所有孩子、我妻子都会经过的地方,上面写着”相信自己”。每天早上霓虹灯亮起来,就是”相信自己”。后来我有机会在英国制作了一枚家族纹章,我的家族是英国老家族了。纹章上刻的就是”相信自己”。我经常跟孩子们谈这个,尤其是他们。我真的觉得,当你面对同辈压力的时候——“你信任谁?他们还是你自己?“当你陷入大量自我怀疑和不确定的时候,你必须相信自己。我觉得我取得的大部分成功——我想这对大多数去创业公司并成功的人来说大概都一样——就是”我对自己做出了好决策有一种信念”。这其实呼应了我之前说的那个关于信念的观点:你怎么做动态消息(News Feed)或者做有争议的事情,怎么做大规模的高成本变革——靠的就是信念。你必须相信,你的眼睛、耳朵和智识结合起来,赋予了你一个具有内在价值的、值得你尊重的观点,而不是读到一篇关于你公司的新闻报道就去相信它,而不相信你自己亲眼所见、亲耳所闻。这就是我们的座右铭。
Lenny: 而且我觉得有一点也很重要,就是你不一定总是对的,这没关系。
Andrew ‘Boz’ Bosworth: 完全正确。相信自己也意味着去冒险,因为你相信自己能够应对冒险失败后的结果。
Lenny: 说得太好了。最后一个问题。我知道你是一名业余摄影师,也许是半专业的,拍了很多旅行摄影作品。
Andrew ‘Boz’ Bosworth: 业余的。业余的,业余的。
Lenny: 业余的。好。你还运营着一个我偶然发现的网站,不知道大家是否知道它,名字很搞笑。我不提了,不知道你是否希望大家……
Andrew ‘Boz’ Bosworth: Warden Shortbow,这是我名字的字母重排(anagram)。Warden Shortbow,Andrew Bosworth 的字母重排,wardenshortbow.com。我热爱摄影。这是真正的热爱。
Lenny: 那么问题是:你拍过的最喜欢的照片是哪一张?
关于艺术与摄影
Andrew ‘Boz’ Bosworth: 顺便说一下,艺术其实是一个很好的话题来谈论”相信自己”。我知道这听起来很俗套,但我认为 Rick Rubin 最近关于艺术是什么以及人们如何创作艺术的访谈说得非常到位。你必须为自己而创作,你必须热爱它。如果其他人也热爱它,它找到了更广泛的共鸣,那当然很棒,但这不是你创作的初衷。如果你开始试图为更广泛的共鸣而创作,那你其实是在追逐别的东西。那是媒体,是娱乐,但不是艺术。那是另一种东西。我说这么多其实是在绕弯子——我确实很喜欢自己的很多照片,很难选出一张最喜欢的。此刻浮现在我脑海中的,一张更有情感冲击力的照片,是我拍的儿子在街上玩耍,穿着雨靴跳进水洼里,穿着那种雨衣之类的东西。那张照片不够锐利,对焦也不实。它是一幅速写,一个意象。
我确实认为 Ansel Adams 曾多次谈到,摄影师的目标是创造一幅作品,让观者感受到当时在场的感受。人们常常忘记,他在暗房方面的造诣甚至可能超过拍摄本身——印放才是他做出惊人成就的地方。我有幸去过他在 Big Sur 的暗房,和他的儿子一起待过,看着他们在那个房间里冲洗照片。他有一套精细的方案,如何在照片的不同部分进行提亮、减光和加深,以获得他想要的质感。他曾为摄影被承认为一种艺术媒介而抗争,当时它并不被认为是艺术。我觉得这与当今关于 AI 艺术的讨论产生了强烈的共鸣——我们又一次试图为”什么是艺术”设门槛,但遗憾的是,你没有资格这么做。
所以这张我儿子的照片,没人会说它是一件技术上的杰作,但作为一个速写,它对我个人而言,也我认为对父母们普遍而言,捕捉到了那些与孩子们共度的、极其动人又迷人的、稍纵即逝的人间瞬间——这就是我想到的那一张。
Lenny: 太棒了。我们会试着找到这张照片并把链接分享给大家。关于 Rick Rubin 这个话题,他说过一句类似的话——他把艺术看作你的日记。我只是在描述我觉得有趣和重要的东西,没有人可以跑来对我说我的日记是错的。这是我的日记,这就是我看待世界的方式,这就够了。最好的艺术正是源自这种信念的鼓舞……他有一段视频说的正是这些,我最近刚看过,非常精彩。Boz,这次对话太愉快了。非常感谢你抽出时间。我期待我们将来一起做一档关于亲子关系和情感关系的播客。开玩笑的,也不是完全开玩笑。最后两个问题:如果大家想关注你的动态,在网上哪里可以找到你?听众怎样能帮到你?
在哪里找到 Boz
Andrew ‘Boz’ Bosworth: 好的。我在 Instagram 和 Threads 上的账号是 @boztank,在 X 上也是,facebook.com/boz。我还有自己的播客,是技术深度访谈类的,所以我可以说它和这个很不一样。每次会深入探讨一两个话题,叫 Boz to the Future。你可以在 Spotify 或 iTunes 上找到它。
Lenny: Boz to the Future,去买点 Quest 的产品吧。
Andrew ‘Boz’ Bosworth: 给自己买个 Quest 3 吧,说真的。兄弟,犒劳一下自己。
Lenny: 没错。或者那款 Ray-Ban 眼镜也是,我是忠实粉丝。Boz,再次感谢你的到来。
Andrew ‘Boz’ Bosworth: 干杯。谢谢,兄弟。
Lenny: 大家再见。非常感谢收听。如果你觉得这期节目有价值,可以在 Apple Podcasts、Spotify 或你喜欢的播客应用上订阅。另外,也请考虑给我们评分或留下评论,这对其他听众发现这档播客真的很有帮助。你可以在 lennyspodcast.com 找到所有往期节目或了解更多关于这档节目的信息。下期再见。
术语表
| 原文 | 中文 |
|---|---|
| 4-H | 4-H(美国青少年发展组织) |
| Ami Vora | Ami Vora(Facebook/Meta 资深产品经理) |
| anagram | 字母重排(anagram,将原词字母重新排列组成新词) |
| Ansel Adams | Ansel Adams(美国著名风光摄影大师,以黑白风景摄影闻名) |
| Atlas | Atlas(希腊神话中擎天的泰坦神,此处比喻承担重任的人) |
| ATT | ATT(App Tracking Transparency,Apple 的应用追踪透明度隐私政策) |
| Be Kind | ”要友善”(Be Kind) |
| Big Sur | Big Sur(加州海岸著名风景区) |
| binary encoding | 二进制编码(binary encoding) |
| Bootcamp | Bootcamp(Facebook 新员工培训项目) |
| Boz to the Future | Boz to the Future(Andrew Bosworth 的个人播客节目) |
| Brian Chesky | Brian Chesky(Airbnb 联合创始人兼 CEO) |
| Chris Cox | Chris Cox(Facebook 早期核心工程师、高管) |
| Communication is the Job | ”沟通本身就是工作”(Communication is the Job) |
| crest | 纹章(crest,家族徽章) |
| Dave Federman | Dave Federman(Facebook 早期工程师,Boz 的好友) |
| Dave Filoni | Dave Filoni(《曼达洛人》主创之一,Star Wars 资深创作者) |
| Dr. Becky | Dr. Becky(临床心理学家 Becky Kennedy,育儿专家) |
| Eric Schmidt | Eric Schmidt(Google 前 CEO) |
| extreme ownership | 极致责任(extreme ownership,源自美军特种部队的领导力理念) |
| eye of Sauron | 索伦之眼(eye of Sauron) |
| eye relief | 眼距(eye relief,镜片到眼球的距离) |
| Failure Corner | 失败角(Failure Corner,播客中的固定环节) |
| FAIR | FAIR(Fundamental AI Research,Meta 的 AI 基础研究实验室) |
| field of view | 视场角(field of view) |
| Galactic Star Cruiser | 银河星舰(Galactic Star Cruiser,迪士尼星球大战主题沉浸式酒店体验) |
| Gell-Mann Amnesia | Gell-Mann 失忆症(Gell-Mann Amnesia,指对媒体在陌生领域深信不疑、在自身专业领域却深知其谬误的认知偏差) |
| Good Inside | 《Good Inside》(Dr. Becky Kennedy 著,育儿与情绪管理书籍) |
| green belt | 绿带(跆拳道段位) |
| hand tracking | 手部追踪(hand tracking) |
| human in the loop | 人在回路中(human in the loop,人类参与 AI 决策循环的模式) |
| IC | IC(Individual Contributor,个人贡献者,非管理岗员工) |
| identity threat | 身份威胁(identity threat,指自我认知受到质疑时的心理防御反应) |
| J.C.R Licklider | J.C.R Licklider(美国计算机科学家,“人在回路中计算”理念的先驱) |
| Jon Favreau | Jon Favreau(《曼达洛人》主创,著名导演/制片人) |
| JV | 二队(Junior Varsity,校队预备队) |
| Karate Kid | 《龙威小子》(Karate Kid,1984 年经典电影) |
| like button | 点赞按钮(like button) |
| Listening is the Job | ”倾听本身就是工作”(Listening is the Job) |
| LLaMA 2 | LLaMA 2(Meta 发布的开源大语言模型) |
| LLM | LLM(大语言模型,Large Language Model) |
| Lou Holtz | Lou Holtz(美国传奇大学橄榄球教练) |
| Mandalorian | 《曼达洛人》(Star Wars 系列剧集) |
| Mark Sleek | Mark Sleek(Facebook 早期工程师) |
| Matthew Ball | Matthew Ball(知名科技与娱乐行业分析师) |
| mental model | 心智模型(mental model) |
| Meta | Meta(公司名,原 Facebook) |
| Michael Crichton | Michael Crichton(《侏罗纪公园》作者、著名作家) |
| mixed reality | 混合现实(mixed reality) |
| monotype | monotype(等宽编程字体) |
| motion blur | 运动模糊(motion blur) |
| multimodality | 多模态沟通(multimodality,使用多种沟通方式和媒介) |
| News Feed | 动态消息(News Feed) |
| pass-through | 透视(pass-through,头显中看到外部真实画面的技术) |
| Patrick Stewart | Patrick Stewart(英国著名演员,《星际迷航》《X战警》系列等) |
| prom | prom 舞会(美国高中/大学正式舞会) |
| Rick Rubin | Rick Rubin(美国传奇音乐制作人,近年出版关于创意与艺术的著作) |
| RPC | 远程过程调用(Remote Procedure Call) |
| SAV | SAV(系统异常值/问题检测) |
| Scott Trowbridge | Scott Trowbridge(迪士尼幻想工程部高管,负责星球大战主题园区) |
| Sheryl Sandberg | Sheryl Sandberg(Facebook 前 COO) |
| Shire | 夏尔(《魔戒》中霍比特人的家园) |
| State of the Union | 国情咨文(State of the Union,美国总统年度演讲) |
| The Dream Machine | 《The Dream Machine》(M. Mitchell Waldrop 著,关于现代计算机发展史的书籍) |
| Thrift | Thrift(Meta 开源的跨语言 RPC 框架) |
| Tracy Emin | Tracy Emin(英国著名当代艺术家,以霓虹灯文字装置艺术闻名) |
| Veritas | Veritas(拉丁语”真理”,亦为哈佛大学校训) |
| wall posts | 留言板帖子(wall posts) |
| 沃伦·巴菲特 | Warren Buffett 的公认中文译名 |
此文档由 AI 分片翻译(translate_long_document)