亨利·埃伦博根 - 人与机器 - [像最佳投资者一样投资，第452集]

我认为他们在自己的领域非常出色。

I think they're great at what they do.

我确实相信在那里工作的人非常有才华。

And I actually believe the people who work there are very talented.

我们的出发点是，这些人都极其优秀，擅长自己的工作，且素质很高。

And we start from the view that these are exceptionally talented people who are actually very good at what they do and are high quality people.

但我们提出的问题是：他们究竟做的是什么？

But what we have said is, what do they do?

他们的局限性在哪里？

And what is the limitation?

其中一个局限是，如果你在一家每天都会严格衡量你风险的公司工作，一旦你连续一个月表现不佳，尤其是三个月内表现糟糕，你的资本就会被削减，很可能被解雇。

And one of the limitations is if you work at a firm that deeply measures your risk every day, And then if you have a bad period of time measured by a month, but certainly three months, you get your capital cut back, there's a good chance you get let go.

这意味着你可能无法拥有比你的职业生涯更长的时间视野。

It probably means you can't have a time horizon longer than your career horizon.

我们还注意到，因为我们不仅通过轶事了解这些情况，还对它们进行了研究。

What we also have noticed, because we not only understand things anecdotally, we also study them.

如果你看一下上一个公开市场的收益季，也就是今年早些时候公司公布的第二季度财报，研究收益波动性就会发现，这是自金融危机以来最波动的一次收益季。

If you look at the last public market earnings season, which was companies reporting q two this past year, if you study earnings volatility, it was more volatile than any earning season since the financial crisis.

尽管在金融危机期间，正如我们都知道的，美国的金融体系本身受到质疑，这意味着经济和市场出现了前所未有的广泛机会分化。

Even though during the financial crisis, as we all know, the fundamental banking system in The US was under question, which meant the economy and the markets as we know it really had a wide dispersion of opportunity.

当市场处于低点和高点时，通常波动性会更大。

And the markets tend to be a lot more volatile when they're making lows and making highs.

然而，这次却是收益波动最剧烈的一季。

And yet, this was the most volatile earning season.

我认为原因很简单：我们估计，80%到90%的机构资金流是由那些以一个月或三个月为考核周期的机构，或者那些必须将这些价格信号纳入考量并优化模型的量化基金驱动的。

I think the reason for that is just basically the fact that we estimate somewhere between 8090% of the institutional flow is driven either by the firms that have one month and three month agency, or the quants that have to take these price signals into account, and then their models are optimized for this.

因此，我们在Durable所提出的观点非常简单：少做一点，才能做得更好。

And so what we have said at Durable is really simple, let's go do less so we can do more.

因为如果我们接受股票的波动性，就必须真正理解业务和人，就像我们在Colliers所做的那样——如果Colliers下跌是因为人们担心商业经纪业务受利率影响，那么市场很可能才是对的。

Because if we're gonna accept volatility in stocks, we have to really understand the business and the people like we do at Colliers, such that if Colliers is down because people are worried about commercial brokerage, because of interest rates, actually the markets are probably right.

他们90%的时间都是对的。

They're probably right 90% of the time.

但如果我们理解这种文化的独特之处，以及它们如何配置资本，我们就能深刻理解Harrison Street的质量，因为我们研究了它二十年，了解运营它的那些人。

But if we understand what's unique about that culture, how they allocate capital, we understand deeply how the quality of that Harrison Street is because we've studied it for twenty years and we know the people who run it.

这正是我们愿意在压力中加码的原因。

Well that's a reason why we're willing to lean into that stress.

当我审视我们的早期成长型投资组合时，情况也是如此，我之前提到过Duolingo。

The same thing when I look at our early stage growth portfolio, I spoke about Duolingo earlier.

它们上次在资本市场对公司的开放期上市，那是2021年。

They came public the last time the capital markets were really open to companies, which was 2021.

关于2021年上市的这批公司，已经有很多文章讨论了它们是在零利率环境下上市的。

And there's been a lot written about the 2021 IPO class, how they came public in a zero interest rate environment.

因此，许多这些公司实际上很难实现盈利，而当2022年利率上升时，很多人正确地认为，干脆把它们全部抛掉。

And so many of these companies actually were gonna have a tough time getting to profitability, and as interest rates went up in '22, I think correctly, a lot of people said, actually, just throw them all out.

2022年，罗素2000成长指数中亏损公司的平均股价下跌了超过70%。

The average loss making company in 2022 in the Russell two thousand growth went down over 70%.

我们的观点是：这很合理，市场可能确实是正确的。

Our view was, makes sense, the market's probably right.

但并不是这些公司都无法适应。

But not all of these can't adapt.

并不是这些公司都没有足够好的业务来实现真正的经济回报。

Not all of these don't have businesses that are good enough to make real economic returns.

我们的观点是，这些公司中并非都没有足够的纪律和组织韧性，在利率变化、要求盈利的新环境中实现转型并取得成功。

Our view is not all of these companies didn't have the discipline and the organizational fortitude to transition and become successful companies in a world that actually required profitability based on the change of interest rates.

Duolingo 就是我们实际上在 2022 年增持的一家公司。

Duolingo is an example of a company that we actually bought more of in '22.

这是一个我们通过美元成本平均法逐步加仓的早期成长型公司例子。

And there's an example of an early stage growth company that we dollar cost average down.

我认为我们有优势，因为如果你的投资是基于规则的，或者只关注一到三个月的时间框架，你就无法持有 Duolingo。

I think we're advantaged in because if you're rule based in what you do or you have a one or three month time frame, you just can't own Duolingo.

你也不可能持有更多的科利尔公司。

You can't own more Colliers.

你提到了人和变化。

You said people and change.

我们稍后再详细讨论

We'll talk more

关于这两点。

about both.

但先说变化，我们正身处可能是我们这一代人所经历过的最大技术变革之中。

But starting with change, we're in the midst of probably the biggest technology shift or change maybe that any of us will ever see.

你曾经投资并亲历过几次其他变革：互联网、移动、云计算。

You've invested and lived through several others, Internet, mobile, cloud.

你能将这次变革与你曾经经历过的其他变革做个对比吗？

Can you put this one in frame of reference with the other ones that you've lived through?

你多次提到研究过去、研究历史并寻找模式。

And you mentioned so many times studying the past and studying history and seeing patterns.

这次有哪些模式可能仍然适用？又有哪些模式需要抛开，从第一性原理重新审视？

What patterns do you think might apply this time, and what patterns do you need to throw out and re underwrite from first principles here?

所以我在2022年的股东信中就开始撰写关于人工智能的内容。

So I started writing about AI in our shareholder letters in '22.

而且，你知道，当时我说过，在见证了互联网、云计算和移动技术之后，我认为这至少会和互联网一样强大。

And, you know, and at the time, I said, having seen Internet, cloud, and mobile, this is gonna be at least as powerful as Internet.

我还说过，我们将以谦逊的态度对待这件事，花时间与最接近它的人交流，并持续学习。

And I said, we're gonna go approach this with humility, go spend time with the people who are closest to it, and constantly be learning.

我也说过，Durable 不是一个主题投资公司，也不是一个复利投资公司。

And I also said durable is not a thematic investing firm or a compounding investing firm.

所以，仅仅因为我们认为某件事会很重要，并不意味着我们的投资会议就会变成‘AI将会很重要’这样的主题。

So just because we think something's gonna be big doesn't mean our investment meetings aren't gonna be AI is going be big.

我们去收购人工智能公司吧。

Let's go buy AI companies.

关键是要真正理解变革，然后理解我们所投资的公司将如何从中受益、如何因此变得更好，或者至少不被颠覆。

It's going to be let's really understand change and then understand how the companies that we invest in are going to benefit from it or become better by it or at least not get disrupted.

我个人认为，它可能比互联网的影响更为深远，我稍后会详细阐述这一点。

I personally, and I'll lay out in a second, think it probably is more impactful than the Internet was.

它不仅会影响每一家科技公司——这在市场上已经有所体现，而且我认为，它还将影响到几乎所有需要白领员工和知识产权员工来推动工作的公司。

It's not only gonna affect every technology company which you see in the markets, but it's also gonna impact in this case, I think almost every company that needs white collar employee and IP employee to drive their work.

所以让我先说第二个，因为我觉得很少有人讨论过这一点。

So let me start with the second one first because I think less people have talked about that one.

作为一名商业研究者，我认为到二月底时，每个人都明白，如果你是一家产品型公司，你就必须了解你的中国成本。

As a student of business, I think that by the end of the February, everyone knew that if you were a product based business, you needed to understand your China cost.

这成了每本商业杂志和所有流行杂志的封面话题。

It was on the cover of every business magazine and then every popular magazine in the tens.

这意味着在全球供应链中，如果世界上某个地区——比如中国和远东地区——拥有巨大的规模和资源，能够以更低的成本生产产品并运送到你的工厂或消费者手中，你就必须了解这一点。

And all that meant was in a global supply chain, if there was a part of the world with huge scale and resources like China and the Far East that could make product substantially cheaper when you landed it at your factory or to the consumer, you had to understand it.

如果你不利用这一点，你就会被淘汰。

And if you didn't leverage it yourself, you were gonna go out of business.

这是第一层衍生影响。

That was the first derivative.

第二层衍生影响是，实际上有很多业务是分布式的。

The second derivative is actually there's a lot of businesses that are spread businesses.

你可以把很多分销业务作为例子来看。

You look at a lot of distribution businesses as an example.

如果你从事分销业务，比如暖通空调行业，这个行业通常会在原材料上加价。

If you distribute talking about a great business like HVAC, The industry tends to put a spread on the raw material.

如果你的暖通空调设备价格上浮3%到5%，那是好事。

If your HVAC unit basically inflates at 3% to 5%, that's good.

如果你的暖通空调设备价格下跌3%到5%，那就是坏事。

And if your HVAC deflates at three percent to 5%, that's bad.

因此，如果你本身是做加价分销的，这就成了问题。

And so if you were a spread business on top of it, that was problematic.

每个以产品为基础或衍生产品为基础的公司都必须了解中国成本。

Every company that was product based or derivative product based had to understand China cost.

我认为同样的道理也适用于人工智能。

And I think the same thing applies to AI.

但我觉得这次不是基于产品，而是基于知识产权。

But I think it's not product based this time, it's IP based.

让我们以马克斯为例来说明。

So let's go back to Max as an example.

一家大多数人会想到的公司，而且确实如此，是一家金融科技公司，以安全且高度合规的方式帮助人们获得信贷，没有任何套路。

A firm most people would think about, and correctly so, as a fintech company that empowers people to get access to credit in a safe way and actually pretty compliant friendly way with no tricks.

而他现在做得甚至更多了。

And he's doing even more than that now.

但归根结底，由于他受到监管，公司承担了大量法律成本。

But at the end of the day, because he is regulated, he has a lot of legal costs in the company.

他与成千上万的商家签订了合同。

He's got hundreds of thousands of contracts with merchants.

他必须监控合作伙伴如何传达信贷信息。

He's got to monitor his partners on how they communicate credit.

而这需要大量的人力。

And that just requires a lot of people.

如果你和他交谈，他曾经公开谈论过这一点，他坚信一家公司可以在合理的时间内保持目前的增长速度。

If you talk to him, and he's talked publicly about this, he's got great belief that a firm can grow at the rates it's growing at for a reasonable period of time.

此外，他们可以在不增加员工人数的情况下实现这一点。

In addition, they can do it without adding headcount.

之所以如此，显而易见，他将借助AI大幅精简过去无法实现的诸多流程。

And the reason for that, obviously he's gonna go lean out a lot of processes that were not possible to go do before AI.

我给你讲个关于这个的有趣故事。

I'll tell you a really funny story with that.

所以，在他公布财报后，我有一天跟他聊了聊。

So I talked to him one day after he reports earnings.

他跟我解释说：你总是问我如何用AI提高效率。

And he explains to me, you're always asking me about how I'm using AI to become more efficient.

我觉得自己有点落后了，但终于搞明白了。

I feel like I was late to the curve, but I finally figured it out.

于是他说，我有一个团队，专门走访公司各部门，了解各项流程。

And so he says, I have this team that goes around the company and understands processes.

我们不是从成本角度出发，而是从精简流程的角度入手。

And we start from not a cost standpoint, but we start from a leaning out standpoint.

我觉得在这方面我确实取得了实质性进展，这就是我能够公开做出这一声明的原因。

And I think I'm making real progress there and that's why I'm able to make this public pronouncement.

他问：这难道不棒吗？

He's like, isn't this great?

我对马克斯说：你为什么不来华盛顿，我们一起去见米奇·拉尔斯？

I say to Max, why don't you come to DC and let's go see Mitch Rails?

因为丹纳赫——他和他兄弟共同创立并担任董事长的公司——已经这样做了四十年。

Because Danaher, who he and his brother started and he's chairman of, has been doing this for forty years.

这叫DBS。

It's called DBS.

他们基本上把精益改善（Kaizen）带回了美国，通过进入工厂、在白板上绘制流程、研究如何精简流程、优化流程，并在一个月后回来确保变革得以持续，因为变革很难维持。

And they basically brought Kaizen back to The US and they've started they're obviously more of a healthcare company now by going into factories, putting processes up in whiteboards, studying how they could lean them out, leading them out and coming back in a month later to make sure because change is hard that the change is held.

然后他们建立了一整套业务体系，这套体系不仅帮助打造了丹纳赫，还培养了超过一打美国《财富》500强的首席执行官，包括刚刚成功扭转通用电气局面的那位高管。

And then they basically built a whole business system, which is basically not only helped build Danaher, but there's over a dozen Fortune 500 CEOs in The United States who started their jobs at Danaher, including the guy who just have turned around GE.

所以既然你对这个如此兴奋，我们不如去拜访美国的教父级人物。

So since you're so excited about this, let's actually go to the godfather of The United States.

我这么说是因为这实在太深刻了，甚至让人难以理解。

The reason I say that is this is just so profound, it's even hard to get your head around.

米奇会说，四十年来，因为我们在中国已经能够大幅优化基于产品的业务和营运资本。

Mitch would say for forty years, because at China, we've been able to really lean out product based businesses, working capital.

但在很多方面，我觉得我们在人类执行的流程上才刚刚开始。

But in many ways, I feel like we're just getting started on processes that are done by humans.

我接下来要谈的第二个例子，是我在这档节目中较少提及的话题：当我第一次真正理解互联网的发展时，我管理着一支全球TMT基金，最大的投资就是亚马逊。

The second example that I'll talk about, and I'm talking about things that haven't been talked about as much on this show, when I first really understood what was going on the Internet, I ran a global TMT fund, my largest investment was Amazon.

我们投资了这家公司。

We invested in that one.

它当时是一家价值一百亿美元的公司。

It was a $10,000,000,000 company.

所以我每年都会去西雅图两次。

So I used to go to Seattle twice a year.

有趣的是，你记得的那些事情，因为西雅图离巴尔的摩很远。

So interesting with the things you remember because it was far from Baltimore.

没人愿意跟我一起去。

No one would come with me.

那是一家小公司，人们以为我理解它。

And it was a small company and people thought I understood it.

我过去每年会和杰夫·贝佐斯一起吃两次午餐。

And I used to go lunched twice a year with Jeff Bezos.

当时，我所在的公司是他最大的外部股东。

At the time, I worked for the firm that was his largest outside shareholder.

这显然是我为公司所做的研究工作。

And it was obviously my research position for the firm.

从这些会面中，我学到了很多东西。

I learned so many things from those meetings.

但我学到的一点是，最优秀的企业在利用技术时，会以降低成本、推动收入的方式运用它，从而在最终市场中获得30%或更多的新增市场份额。

But one of the things I learned is the very best businesses that leverage technology, leverage it in a way where they use it to lower costs and drive revenue that result in them gaining 30% or more incremental market share in their end market.

然后，它们会将这种单位经济效益优势。

And then they take that unit economic advantage.

重新投资于某种持久的事务。

And they reinvest it in something that is persistent.

即使他们的竞争对手明天醒来，做出完全相同的事情，用和他们一样优秀的人才。

Even if their competition were to wake up tomorrow and do the exact same thing with people just as good as they are.

在我看来，这就是持久竞争优势的定义之一：即使你的竞争对手发起正面攻击，用同样优秀的人才做完全相同的事，也无关紧要，因为你已经领先太多了。

And to me, that's one of the definitions, a durable of a competitive advantage is if your competitor does a competitive mode attack doing the exact same thing with people as well or Doesn't matter because you're too far ahead.

正如我们所有人都逐渐了解到的亚马逊，他们利用了将包裹送达你手中时3%到5%的成本优势，以及在同一个包裹中放入多件商品的能力。

And as we all have come to understand with Amazon, they took that three to 5% cost advantage of getting that box to you and their ability to put more than one item in the box.

他们利用这一经济优势，进一步投资建设实体配送中心，将资金和基础设施投入到持续降低3%至5%成本曲线的行动中，长达二十年之久。

And they use that economic advantage to then go build fulfillment centers that are physical to reinvest into capital and infrastructure that allowed them to go down that three to 5% cost curve for twenty years.

正如我之前在节目中所说，当他们终于醒悟过来，只有那些仍然拥有规模、客户关系和信任的公司——比如沃尔玛和好市多——才能跟上他们的步伐。

As I said earlier in the show, they woke up and the only people who could play their game when eventually everyone realized what they were doing were the people who still had the scale and the customer relationships and the trust of Walmart and Costco.

最终，当他们意识到时，这三家公司都表现得非常出色。

And then eventually when they figured out, all three of them were great.

但问题在于，其余的零售业并不那么出色。

Now the problem is the rest of retail was not so good.

这正是我们在这里所思考的问题。

That's what we think about here.

所以如果我复述一下，我们通过米奇以及其他类似的人看到了持续改进在实体产品领域带来的四十年优势。

So if I say it back to you, we've seen through Mitch and others like him, this forty year benefit of kaizen brought to physical product world.

而人工智能标志着持续改进理念首次进入人类工作领域。

And that AI represents a sort of kickoff of Kaizen to human work world.

这将带来许多类似你刚才提到的亚马逊故事，即有人早早踏上了这样的曲线，别人就再也追不上了。

And that that is gonna have lots of stories like the Amazon story you just said, where someone gets on one of these curves early and they can't be caught.

因此，你肯定在努力寻找那些可能的人是谁。

And so you're trying to I'm sure to find who those people might be.

所以我们正在努力做两件事。

So we're trying to do two things.

我们试图找出那些可能的人，同时确保我们不会被一家公司拖垮，这家公司基于上一代的竞争风险曾是一家优秀企业。

We're trying to find the people who that might be, and then we're trying to make sure we don't Get killed by own a company that based on last generation competitive risk was a great company.

我们曾正确地认为它具有竞争优势，拥有良好的运营文化，由具有主人翁精神的高素质领导者带领。

And correctly, we thought was competitively advantaged and had a good operating culture, was led by high quality people who thought like owners.

但由于颠覆性变革改变了世界，他们未能适应。

But because discontinuous change changed the world, they weren't able to adapt.

理想情况下，我们会去寻找那些已经占据优势的公司，

Now ideally what we go do is we go find those already advantaged companies,

然后利用这一点，让它们从优秀走向卓越。

and then they leverage this to go from good to great.

如果你想想你合作过的那些最擅长应对变革的首席执行官们。

If you think about all the people that have navigated change as CEOs the best that you've worked with.

我与戴夫·杜菲尔德一起投资，他是他最新公司的创始人，他现在正投身于人工智能领域，这个人简直太出色了。

I'm an investor with Dave Duffield and his latest company and so he comes to mind because now he's tackling AI and the guy is incredible.

如果你想想你见过的像米奇、杰夫、戴夫这样的人，他们运作的方式中，最让你印象深刻的是哪些方法？首先是他们自己如何调整心态，然后是他们的团队如何适应这些快速变化的环境？

If you think about whether it's Mitch or Jeff or Dave or people like this that you've seen operate, what methods have impressed you the most of how they themselves adapt, first their own mentality and then their teams to these fast changing circumstances?

这个问题也适用于所有正在经营企业、面临同样变革的人——这种变革既是风险，也是机遇。

And this is a question for everyone out there that's running businesses that is facing this same change that's at risk and an opportunity at the same time.

我们来拿路易斯开刀吧。

Let's go pick on Luis.

我们之前讨论过路易斯。

We talked about Luis earlier.

Duolingo在人工智能方面有很多机会，但也面临很大风险。

Duolingo has a lot of opportunity with AI and a lot of risk.

股价每天都会根据情况波动。

And the stock, depending on the day, reflects it.

当OpenAI演示如何使用其技术进行翻译时，股价常常会下跌。

When OpenAI demos how you can use OpenAI to basically do translation, a lot of times the stock goes down.

或者当苹果展示AirPods如何实现实时物理翻译时。

Or when Apple shows you how AirPods can be used to in in the physical Live translation.

翻译。

Translation.

股价就会下跌。

Stuff goes down.

实际上，我认为市场定价并没有错。

And actually, don't think the marker's wrong.

现在可能在幅度上有所偏差，但我认为市场想表达的是：存在风险。

Now it's probably wrong in the magnitude, but I think what the market is saying is There's a risk.

这里存在风险。

There's a risk here.

因此，在我们的投资组合中，这可能是风险最高的股票之一。

So that's probably in our portfolio one of the higher risk names.

但当路易斯、马克斯或戴夫·达菲德等人出现时，我们关注的是什么？

But what do we look for when Luis or Max or Dave Duffields?

我认为我们首先关注的是一个已经运营良好的企业。

I think the first thing we look about is a business that already is operating well.

因为如果你在面对变革时运营状况不佳，你就无法同时做好两件事。

Because if you're not operating well when you have to deal with change, you're not gonna be able to do two things at once.

你无法在扭亏为盈的同时还表现良好。

You can't do a turnaround and do well.

我认为第二点是，一个已经在其初始市场中取得胜利的企业。

I think the second thing we would say is a business that already was winning in its first end market.

我们对‘胜利’的所有定义，都是指显著获得市场份额，创造真实的经济利润，从而能够投资于下一个增长曲线。

All the definitions we would have about winning, substantially gaining market share, driving real economic profit that allows it to reinvest in this next s curve.

我们非常关注的另一点是人。

And also the other thing we really care about is people.

我们在Durable非常强调韧性。

We really stress resiliency at Durable.

这正是我们如此喜欢投资于主动管理者的原因之一，我们可以去研究他们的韧性。

It's one of the reasons we like so much investing in active managers, We can go study their resiliency.

当你作为Max的投资者，看到他在Slide时展现出的韧性，现在你明白他必须具备韧性，才能在竞争对手之前实施AI以降低成本并推动收入增长时，你会说：我们已经见过他在压力下的表现了。

When you're an investor in Max and you saw how resilient he was in Slide, and now you understand how he's gotta be resilient to implement AI to lean out his costs and drive his revenue before his competition does, you're like, we've seen him under stress before.

我认为你寻找的是那些有见解、能执行，但同时也保持谦逊的人。

I think you're looking for people who have a perspective, can execute, but also are humble.

因为我们在Durable每六个月都会写下我们对AI的看法并进行更新。

Because we at Durable write down our views on AI every six months and we update them.

随着我们进入这个变革时期，我们的错误可能更少了。

And as we've gone more into this period of change, probably we're less wrong.

与我们每六个月更新一次观点时相比，我们的观点现在更加有依据了。

Our perspective is more informed you would say than it was as we change it every six months.

但我们正在学习。

But we're learning.

我们有幸从事一份允许我们花时间阅读、思考和与聪明人交流的工作。

We have the benefit of having a job that allows us to spend our time reading and thinking and talking to smart people.

所以即使这些人非常有才华，他们很可能没有我们这么多时间去深入研究，因此我们必须在方法上保持谦逊。

So even if these are really talented people, most likely they don't have as much time to go do it as we do, and we have to be very humble in our approach.

因此，他们必须有自己的见解，弄清楚如何付诸实践，不被吓倒，但同时也必须保持谦逊并持续学习。

So they have to have a perspective, figure out how to go do this, not be paralyzed, but they also have to be humble and constantly learning.

最后，我认为作为投资者，我们在实践中必须考虑的是‘背书备忘录’。

And the last thing I think practically we have to think about as investors is backs to the memo.

如果你面临高风险情境——我们都同意多邻国和富达是这样的情况——而且你非常接近这一变革阶段，那么你就需要为这种风险获得补偿。

If you have a high risk situation, which we would all agree Duolingo is and firm is, and you're really close to that part of the change, then you need to be compensated for the risk.

因此，我们会告诉多邻国的人：是的，由于这种风险，折现率已经上升。

And so what we would tell people on Duolingo is, yeah, because of this risk, the discount rate has gone up.

但相应的，机会可能也增加了，我们可以清晰地阐述这一点。

But probably commensurate the opportunity has also gone up, we can articulate it.

路易斯曾谈到，他让效率提升了20倍，并生成了内容。

Luis has talked about being 20 x faster and generating content.

我们已经看到了。

We've already seen it.

他公开表示自己开发了国际象棋产品，这款产品表现极为出色。

He's publicly said he developed chess, which is doing incredibly well.

我的孩子们非常喜欢这个产品。

And my kids really love that product.

我的意思是，他们上瘾了。

I mean, they're addicted to it.

起初，只有两个人花了六个月时间，后来他又增加了四个人，九个月内就开发出了这款产品。

First, it was two people for six months, and then he added another four people, and he developed a product in nine months.

这是他迄今为止最成功的产品。

That's the best product he's ever done.

如果你问他，过去做这样的产品需要多久，他会说，我不知道。

And if you ever talk to him how long would it have taken in the past, he's like, I don't know.

我可能需要四到六倍的人手，而且他们花的时间会是四倍长。

I probably need four to six x as many people, and it would have taken them four times as long.

所以你可能会说初创公司也能做到，但他真的做到了。

And so you could say a startup could do that, but he's doing it.

他是在自己的平台上完成的。

He's doing it on his platform.

这款产品的日活跃用户数已远超百万，且增长速度惊人，这是一个需求强烈但服务不足的细分市场。

That product is well over a million DAUs and is growing at astronomical rates, and it's a great underserved end market.

这可能成为一个巨大的生意。

It could be a huge business.

所以，是的，折现率上升了，但这家公司从诞生之初就是为AI打造的。

And so, yes, the discount rate has gone up, but this company was purpose built for AI.

你确实有一位研究过AI并在卡内基梅隆大学教授过AI的人，他领导着一支敏捷的团队，为人谦逊，持续学习，这正是他比其他人更快的有力证明。

You actually have a person who studied AI and taught it at Carnegie Mellon and has an organization of a players who are agile in his company and is humble and constantly learning a proof point on how it's making him faster than other people.

它正在创造真正的价值。

It's driving real value.

显然，这意味着它从一个点解决方案或几个产品发展为一套完整解决方案的概率更高。

And, obviously, that means the probability of it going from a point solution or a couple products to a suite is higher.

因此，这就是我们所寻找的。

And so that's what we look for.

我可以把您提到的几个最让我感兴趣的观点先堆叠一下，然后再提一个更大的问题吗？

Can I stack a couple of the things that you've said that interest me the most until a bigger question?

如果我把物理改善和数字改善这两个概念简化一下，它们如果生个孩子，可能就是机器人。

If I take physical kaizen and digital kaizen just to shrink the concepts down, if those were to have a kid, it might be robotics.

因此，我非常好奇，您如何看待这种可能性：我们可能会迎来第二波物理劳动力经济浪潮，其规模远超数字劳动力经济，未来四十年内可能会出现第二波改善应用。

And so I'm really curious how you approach the potential with an unknown timeline that we might get a second wave of the physical labor economies way bigger than the digital labor economy, that we may get a second application of Kaizen over the next forty years.

我们之前看到的第一波是由米奇及其公司推动的。

The first one we saw from Mitch and company.

您如何看待这一机遇和潜在变革？

How do you think about that opportunity and potential change?

这是我们认真思考过的问题。

It's something we have really thought hard about.

这是我们尝试过的方法。

Here's what we have tried to do.

首先，我们通过与创业者会面，并与我们参与且了解的、在这一领域领先的公司交流，试图学习更多。

First of all, we have tried to get smart by meeting with the entrepreneurs and talking to the companies that we're involved in that we know that are leading this area, just to try to learn.

我们最近才开始在机器人领域这样做。

We only recently did this with robotics.

因此，如果我们一开始就把我们的结论写下来，并且谦逊地承认这些结论在人工智能领域每六个月都可能改变，或者在你所说的更多数据业务或数字业务中发生变化，那么我们实际上只是上个月才第一次系统地记录下这些内容。

So if we started writing down our conclusions, and then being humble that it could change every six months in AI, in what you would say is more data businesses or digital businesses, we only started doing this literally in the last month where we documented it for the first time.

我要做一件我不太喜欢的事，那就是我知道我们在这里的观点还非常早期，很可能完全错误。

I'm gonna do something that I don't love to do, which I know our views here are very early and probably deeply wrong.

但我还是向你们分享我们的初步结论：在某些应用场景中，已经很明显，成本低于同等的模拟处理器或物理劳动流程。

But I'll give you our initial conclusions, which is in certain use cases, it's pretty clear that already the cost is lower than the equivalent analog processor Physical labor process.

物理劳动流程。

Physical labor process.

然而，正如我们都知道的，这正是机器人技术最初始、最糟糕的阶段。

And yet, as we all know, this is the earliest and worst robotics gonna be.

由于机器之间相互迭代，并由通用模型而非专用模型驱动，这正乘上一条明确呈几何级数增长的曲线。

And because machines are iterating with machines and is being powered by general purpose models, not by specific purpose models, this is riding a curve that is definitely geometric.

因此，回到亚马逊这个塑造了我们对持久性思考的思维模型。

And so back to this mental model of Amazon that drives so much of what we think about durability.

亚马逊能够做到的是，连续二十年以每年3%到5%的速度降低发货成本。

What Amazon was able to do is ride a cost curve where they were deflating the cost of sending a box out at 3% to 5% a year for twenty straight years.

而那些没有利用正确分销基础设施、没有在当时及时投资机器人技术以及在机器学习模型用于规划库存和管理供应商时未能跟进的企业，其成本曲线可能每年上涨3%到5%，或者至多保持平稳。

And the people who did not leverage the right distribution infrastructure, the right investment in robotics at that time, they bought Kiva, the right ML models when they came into play about how to basically plan inventory and deal with suppliers actually were probably at a curve that probably inflated at 3% to 5%, but at best were flat.

在低利润率的行业中，这种差异如果连续复利五年，老实说，你就只需要明白这一点。

And that differential in a low margin business, you compound it over five years, honestly, that's all you need to know.

我认为，这正是我们正在Durable逐渐理解的：在许多领域，机器人技术已经达到了同等水平。

And that's, I think, what we're starting to get our heads around at Durable, which is if in many areas robotics is at parity.

现在有很多数据表明，在某些情况下，机器人成本更低。

Now there's a lot of data that says it's lower cost in certain cases.

而应用场景即将大幅增加。

And the use cases are about to go up.

而在当前阶段，由于规模扩大、人力投入增加、知识产权的收购，以及你将能够利用这些通用大语言模型来驱动这些现有应用场景，成本下降幅度可能不再只是3%到5%，而是更接近15%到20%。

And the cost on the existing use cases probably don't go down at 3% to 5% at this point in the curve because of the scale brought in, the human capital brought in, the IP bought on, and the fact that you're gonna be able to use these general purpose LLMs to power it, it probably goes down at more like 15 to 20%.

但也许这就像摩尔定律，成本下降得更快。

But maybe it's Moore's Law and it goes down even faster.

那么五年后，我们就会醒悟：那些把自己定位在一条成本曲线上的公司，如果与那些定位在另一条成本曲线上的公司竞争，后者可能成为幂律效应的商业赢家。

Well, then we wake up in five years and the people who put themselves on one cost curve, if they compete against the people who put themselves on the other cost curve, those could be power law businesses.

我们深入思考过的问题是：谁将从这条成本曲线中获益？即使他们的竞争对手明天就醒悟过来，即使他们投入同等的资金，即使他们能招聘到同样优秀的人才——而这几乎不可能，因为他们从未投资于分布基础设施或技术基础设施——这些对手至少落后两到三年。

What we have thought really hard about is who's gonna go benefit from this curve where their competition, even if they woke up tomorrow, even if they put the same amount of money into this problem, even if they could hire the same quality people, which is unlikely, because they have not invested in the distribution infrastructure or the technological infrastructure to compete on this curve at minimum is probably two to three years behind.

而他们每拖延一天，差距就会进一步拉大。

And every day that they wait, they're probably getting further behind.

Durable的每个人都清楚，我们对变革的理解为何能助力我们在Duolingo、Affirm和Shopify上的投资。

It's apparent to everyone at Durable why our understanding of change has been great for our investments in Duolingo and Affirm and Shopify.

但实际上，我们认为这种理解即将带来更大的优势。

But actually, think it's about to be really advantaged.

当我们投资于经济中另外70%的领域时，我们对人性和变革的理解将发挥更大作用。

Our understanding of people and change as we go invest in the other 70% of the economy.

如果你想想你投资过的所有企业，你有没有特别偏爱某种竞争优势或竞争来源，以至于它成为你整个投资流程的核心？

If you think about all the businesses you backed, do you have a favorite kind of competitive advantage of source of competitive advantage so much about your whole processes?

它们已经具备这种优势了吗？

Does it have it already?

它们是否正朝着这个方向发展？

Is it on its trajectory to get there?

有不同类型的优势，比如规模效应、网络效应等等。

There's different kinds, scale, network effects, etcetera.

你是否特别倾向于某些优势，认为它们是长期竞争优势的最佳来源？

Are there favorites that you find yourself returning to as the best sources of long term competitive advantage?

我特别喜欢两种，而且它们实际上非常不同。

I really love two, and they're actually quite different.

我喜欢实体房地产。

I love physical real estate.

我喜欢亚马逊，或者我们投资了Carvana。

I love Amazon or we're investors in Carvana.

我喜欢他们的翻新中心。

I love their reconditioning centers.

因为归根结底

Because at the end of the day

你没法快速搭建起这些设施。

You can't spin those things up.

你没法快速搭建起这些设施。

You can't spin those things up.

这些事情非常复杂，你得先拿下土地，选对位置，建立合适的网络，然后投入合适的资本支出和系统，还要打造正确的运营文化。

These things that are super messy, you gotta acquire the land, you gotta put it in the right place, you gotta build the right network, then you gotta go stand it up with the right CapEx and the right systems, and then you have to have the right operating culture.

这真的非常困难。

This is really, really hard.

如果你把房地产选在错误的地方，运输成本就会更高。

If you put your real estate in the wrong place, then your cost of transport is more expensive.

而你要在那里建立的这种文化，极其难搞。

And this culture you gotta go build there is super hard.

所以我非常喜爱这些存在于现实世界中的护城河。

So I deeply love these physical world moats that exist.

我们的投资组合在某种程度上确实包含了很多这样的护城河。

And really our portfolio has a lot of them in one way or the other.

因此，当你我讨论机器人技术时，我的第一反应就是物流配送。

And that's why when you and I talked about robotics, my mind went to distribution.

问题是，机器人技术能在哪些方面让本已占优势的企业变得更强大？

It's like, where can robotics basically take already advantaged businesses and make them more advantaged?

我深信另一类东西——那些看似抽象却极其困难的要素。

The other thing I really believe is these soft things that are incredibly hard.

我经常想到丹或者她，因为米奇和史蒂夫所做到的事情简直难以想象：近四十年来，他们持续以20%的复合增长率创造财富，而他们所从事的业务并没有像MetaMite或谷歌那样的深层物理护城河，也没有像那些坚信开源AI的人所依赖的数据网络效应，那些数据网络效应确实惊人。

I think about the example of Dan or her so much because what Mitch and Steve have done is stunningly hard to imagine that for nearly forty years, you've compounded wealth at 20% in something that didn't have deep physical moats or didn't have data network effects like MetaMite or Google have or the people who believe deeply in open AI, those data network effects are amazing.

但真正厉害的是那些在人力资源上极为敏锐的人，他们真正理解人才的含义，而不是只看简历上的头衔。

But the ones where you're really sharp on human capital, you're really sharp on what talent really means, not the sticker of talent.

他们对运营卓越、企业文化、持续改进以及背后的系统有着深刻把握，就像卡赞那样。

You're really sharp on your operating excellence, the culture of it, the constant improvement of it, the system behind it, like Kazan.

然后你将资本分配给你的企业，以真正让他们负起责任。

And then you allocate capital against your businesses to really hold them accountable.

我觉得这太棒了。

I think it's amazing.

在我们的投资组合中，即使我带你逐一了解FirstService的业务——Jay Hennick是其董事长，以及Colliers的业务——他是CEO，同时也是这两家公司的最大股东，但它们实际上都没有这种超强的竞争优势。

In our portfolio, even though if I were to walk you through the businesses that FirstService is in which Jay Hennick's chairman of and what Colliers has where he's CEO, he's the largest shareholder of both, none of them actually have these super sharp competitive advantage.

但如果你真正研究过Jay，真正理解他的人才文化，以及他是如何吸引人才并让员工承担责任，还有他如何实现激励机制的去中心化，

But yet if you really have studied Jay and you truly understand his human capital culture and how he basically attracts and hold people accountable and his ability to basically decentralize incentives.

那么你会发现，无论是住宅物业管理、屋顶维修与修复，还是资本配置、出售没有发展前景的业务，以及收购新企业，他都能以一种持续与员工利益保持一致的方式做到这一切，这非常令人印象深刻。

So people are aligned in businesses, everything from residential management of condos to roofing and restoration, and then how they allocate capital, how they sell things that don't have a path to be great, and how they buy businesses, but they do it in a way where it actually aligns incentives in constant with them, it's super impressive.

另一点是，如果我们投资小型公司，这些公司在形成时往往已经相当庞大了。

The other thing is if we're gonna invest in small companies, those companies by the time they form them tend to be pretty large.

我们必须非常敏锐地理解其他类型的竞争优势。

We have to be pretty sharp at really understanding other competitive advantages.

你提到过几次你写的内部备忘录。

You mentioned memo a few times, the memos you write internally.

你认为一份出色的投資備忘錄應該具備怎樣的結構？

What have you learned makes for a fantastic structure of an investment memo?

對你來說，什麼樣的方式最有效？

What works for you?

我們是一個注重書寫的文化。

We're a writing culture.

因為歸根結底，人類本質上就是人類。

Because at the end of the day, human beings are innately human.

當我們參與某件事時，很難保持所需的那種執行層面的距離，來真正對自己的判斷負責，並讓公司對你希望他們做到的事負責。

When we are involved in something, it's very hard to have that executive distance that you need to do to really hold yourself accountable to what you thought, and actually hold the companies accountable to what you would like them to do.

尤其是因為我們確實了解我們投資的人，我們投資的都是高品質、有趣的人，而且我們衷心希望每一個人成功。

Especially since we really do know the people we invest in, and we invest in really high quality interesting people, and we're deeply rooting for everyone to succeed.

與風險投資公司或私募投資公司不同，我們必須能夠識別哪些事情沒有起色，以便及時出售。

Unlike a venture capital firm or private investing firm, we have to be able to understand when things aren't working out so we can sell them.

但同樣重要的是，如果我們要投資像Duolingo這樣具有真正風險的項目，我們必須能夠察覺像Duolingo這樣的事情何時出現轉折點，而這些轉折點可能是別人沒注意到的，這樣我們才能進場加碼。

But just as importantly, if we're gonna go invest in something that has real risk, like Duolingo, we have to understand when things like Duolingo are really inflecting, and maybe we see it other people don't do, so we can go buy more of it.

所以我们必须能够做到这两点。

So we got to be able to do both.

当我们撰写投资备忘录时，它是为我们投资理念服务的，确保我们已经做了充分的研究，并能清晰地阐述这家公司为何具有竞争优势，或未来将具备竞争优势。

When we write an investment memo, it's in service of our investment philosophy, making sure we've done the work and we can clearly articulate why is a company competitively advantaged or will it be?

它需要做到什么？

What would it have to do?

为什么这种运营文化是卓越的？

Why is this operating culture excellent?

或者为什么它具备卓越运营文化的种子？

Or why does it have the seeds of an excellent operating culture?

为什么这位领导者具有所有者思维，能够真正让业务变得更好，而我们定义的‘更好’就是通过周期赢得市场份额。

And why does this leader think like an owner, where they can basically make the business better, which we define as gaining market share through cycle.

我们认为，他们能够有效配置内部和外部资本，从而推动更持久的增长，并随着时间推移提升其资产价值。

And we think that they can allocate internal and external capital, such to both drive more durable growth, and also make their asset base more valuable over time.

这就是我们的投资备忘录。

That's our investment memo.

但同样重要的是，通过建模并明确说明我们所跟踪的内容，我们必须在每季度进行运营回顾时——在Durable，我们会与整个投资团队一起审查投资组合中的每一项投资——来审视这些公司实际的表现与我们原先预期的对比情况。

But then just as importantly, through both modeling, but clearly spelling out what we're tracking, we have to then be able to, when we do our quarterly, what are really our operating reviews at Durable, where we go through the entire portfolio with the entire investment team on every single investment, we have to look how actually the companies are doing against what we thought they would do.

而在Durable，每一项投资，如果我们持有三年，我们都会进行一次三年回顾，看看当初我们所评估的内容与实际结果之间的差异。

And then every single investment at Durable, if we own for three years, we actually do a three year look back on what we underwrote and what it did.

这一点，就像你职业生涯中的许多事情一样，如果你早点知道就好了。

And that one, like so many things in your career you wish you would have known earlier.

投资的美妙之处，以及在我看来优秀的投资者，往往在70岁时比50岁时更出色。

The great thing about investing and the great investors to me actually are better at 70 than they are at 50.

希望我在50多岁时，比30岁做这项工作时做得更好。

And hopefully, I'm in my 50s, I'm better than I was when I did this at 30.

你只是在不断学习。

And you just learn.

你对模式的理解更加深刻。

You understand patterns better.

希望你始终保持谦逊，不至于固步自封。

Hopefully, you remain humble so you don't actually get too stuck in your ways.

你要围绕自己身边聚集更聪明、更优秀的人，无论是内部还是外部。

You surround yourself with smarter, better people both internally and externally.

但其中一点是，你要建立更好的流程。

But one of them is you do better processes.

我们两年前开始实施的一个流程是，我们一直做季度关键绩效指标或运营复盘，但从未回头去看那些我们持有了三年的投资，只是简单地问：当我们做这些复盘时，它们其实非常简单。

And one of the process that we always started two years ago was we always did quarterly KPIs or operating reviews, but we didn't go back and look at an investment that we own for three years and just say and when we do them, they're so simple.

我们会说：三年前，我们认为它们会做到X，而现在它们做到了Y。

We say, three years ago, we thought they would do X and now they did Y.

当然，接下来的讨论会是：哪里不同？为什么？

Now of course the conversation is where was it different and why?

但实际上，这场会议的准备工作——至少在书面层面——非常简单。

But actually, the preparation for that meeting is, at least on the written side, so simple.

只有两页纸。

It's two sides.

但当然，我们都是凡人。

But then, of course, we're all human.

尽管我们努力互相问责，但如果每个季度聚在一起时，事情只是稍有偏差，人们往往会为此找借口。

And even though we try to hold each other accountable, if you get together every quarter and something deviates a little bit, you tend to excuse it.

当然，如果连续12个季度都稍有偏差，实际上，这已经摆在眼前了。这就是为什么我们如此坚信投资备忘录的重要性。

Of course, if it deviates a little bit for 12 straight quarters, actually, staring you in the face, That's why we're such a big believer in investment memos.

我想在2022年，你曾进行了一次巡回交流，与CEO们探讨了

I think in 2022, you went and did a tour talking to CEOs about the state of

市场状况、运营原则以及类似的话题。

the market and operating principles and things like this.

我很好奇想听听你对那次巡回的反思，但更感兴趣的是，如果你今天要对投资组合中的每一位CEO进行一次巡回（也许你正在积极进行），那么今天传达的信息与2022年相比会有何不同？

I'm curious to hear you reflect on that tour, but even more interested to hear if you were to do a tour of every CEO in the portfolio today, and maybe you're doing this actively, what the message would be today that's different than the message was in 2022?

我觉得在2022年，我们确实拥有能为对话增添价值的专业知识。

I felt in 2022, we truly had expertise to add to the conversation.

那就是我们在最高层面上，顺便说一句，我并不是在说‘我们’与‘他们’对立，而是指我们所有人。

And that was we at the highest level, and by the way, I don't say we versus us, we all.

我交谈过的每一位CEO、每一位投资者，甚至包括Durable这家真正重视现金流、从不认为资金会永远免费的基本面投资公司，都基于近十年的免费资金环境做出了简化的假设。

Every CEO I talk to, every investor I talk to, and even Durable who is a fundamental investment firm that really values stuff on cash flow and never felt money was gonna be free forever, had made simplifying assumptions based on almost a decade of free money.

任何接受过企业估值训练的人都明白，最终所有公司都必须基于自由现金流和有机增长来估值。

If anyone who has been taught how to value companies understands at the end of the day, all companies have to be valued on free cash flow and organic growth.

当时，全球高达30%的国库券收益率竟为负值。

At the time we got to a point where 30% of all treasury bills in the world actually had negative yields.

相对于通胀而言，借款人甚至还能获得报酬，这使得风险投资家在估值公司时完全不关心盈利能力变得合乎逻辑。

And relative to inflation, were being paid to borrow, which basically means it was logical for venture capitalists to value companies and not care about profitability at all.

公开市场上的公司也觉得有理由去收购那些永远无法覆盖资本成本的低质量企业，利用廉价债务完成收购。

It was logical for companies in the public markets to buy low quality businesses that could never own their cost of capital but use cheap debt to go do it.

如果你坐在公司董事会里，确实不会去追问增长、盈利能力和创新之间的权衡问题，因为根本没必要。

It was logical why if you sat on companies' boards, you really wouldn't ask hard questions about trading off growth, profitability, and innovation.

因为你根本不必这么做。

Because you didn't have to.

如果我们回顾之前讨论复利时的对话，我们曾分阶段进行过这项研究，观察公开市场，并问自己一个简单的问题。

And if you go back to the conversation we had when we eventually studied compounding, we went back and we ran that study in periods, we looked at the public markets and we asked ourselves a simple question.

在正实际利率的世界里——也就是美国股市历史上除那段短暂时期外的全部时间——平均大约有40家复利型企业；而在资金廉价的时期，这个数字达到了120家。

In the world of positive real rates, which is the entire history of The US equity market except for that short period of time, which I don't know if we'll ever see again, there's on average about 40 compounders and during the period of time of free money, there was 120.

所以这样做起来容易了三倍。

So it was three times easier to do it.

然后我们问了自己一个简单的问题：哪些模式只有在资金免费时才会出现？

And then we asked ourselves a simple question, what patterns only exist when money is free?

因此，毫不奇怪，每个人都认为推动增长和盈利的模式是永恒的，无论资金是否免费都有效。

So not surprisingly, everyone would imagine the pattern of driving growth and profitability is perennial and actually works regardless whether money is free or not.

另一个非常有趣且对我们至关重要的发现是，当我们做这项研究时，如果你是一家小公司，你不需要在GAAP下实现盈利。

The other thing that was really interesting, which was really important to us and gave us confidence to go buy more of the Duolingo's in '22, when we did this work, is that if you're a small company, you don't have to be GAAP profitable.

你甚至不需要让投资回报率高于你的资本成本。

And you don't even have to have an ROI that is above your cost of capital.

但你必须展现出朝着这个目标前进的进展。

But you do have to show progress in your path towards it.

那么，哪些做法行不通呢？

And then what doesn't work?

那就是一家既不产生财务回报，又没有任何进展的公司。

Well, a company that doesn't earn financial returns and is showing no progress.

另一个无效的做法是我们很少做的：以高价收购低回报的业务，同时使用极低成本的债务进行杠杆操作。

And the other one that doesn't work, which we don't do a lot of, was go buy a low quality return business at a high price, but leverage really cheap debt.

因为我之前见过类似的周期，真正具备相关专业知识，而且我们始终作为长期合作伙伴支持所投资的公司，无论它们是私营还是上市，我们都希望帮助它们顺利过渡。

So I felt strongly because we had seen cycles before, and because we truly had expertise, and also because we're really our companies that we invest in a long term partner, we want to invest in companies that are private, instill in them when they're public, we want to help them with transition.

我认为我们拥有许多人都未曾深入探索过的专业见解和独特视角。

I thought we had expertise and a perspective that many people had not gone in before.

因此，为了说明这一点，我们与许多处于这种情况的公司进行了这类对话。

And so to reference what this meant was we had these conversations with a number of our companies that were in this situation.

我们与Toast的Aman、Duolingo的Luis以及Affirm的Max都进行过类似对话。

And so we had some version of this conversation with Aman at Toast, with Luis at Duolingo, with Max at Affirm.

但每一次对话都有所不同。

And all of them were a little different.

比如与Luis，我和他以及他的CFO共进晚餐，他的CFO和他一样非常出色。

As an example with Luis, I had dinner with him and the CFO, and his CFO is very talented just like he is.

我只是向他们展示了这些数据。

And I just presented them the data.

了解他们之后，我知道他们会提出很多问题，而他们确实问了很多很好的问题。

Knowing them, I knew they would have a lot of questions on it, and they asked a lot of good questions.

我跟他们说的另一件事是：听着，Luis，当我们投资你们公司时，我总是会问人们一个问题：我们能为你们提供什么？

The other thing I said to them is, look, Luis, when we invest in your company, one of the things I always ask people is, we'll articulate what we're looking for in you.

但你们在寻找我们提供什么？

What But are you looking for in us?

他提到的一点是，我将是一位首次担任CEO的人。

And one of the things he said is, I'm going to be a first time CEO.

我的感觉是，你们会基于你们的经验看到一些我可能看不到的东西，因为这对你们来说是全新的。

And my sense is you're going to see things based on your experience that maybe I don't see because this is new to me.

我说，我之所以想和你们共进晚餐，并不是因为我拥有所有答案，而是因为我坚信你们在自己的领域里占据主导地位。

And I said, the reason I wanted to have dinner with you is not because I have all the answers, but I have a strong view that you're dominant in what you do.

人工智能对你们来说非常棒。

AI is amazing for you.

当时人工智能才刚刚起步。

AI was just getting started.

你们拥有非常独特的人力资本文化。

You have a very unique human capital culture.

但如果你要向外界传达这种市场优势，并不意味着你必须在30%的长期利润率目标上一步到位。

But if you're going to communicate to people this strength in the market, it doesn't mean you need to get to your long term margin targets at 30%.

但你必须展现出朝着这个目标的进展。

But you have to show progress towards it.

对于其他公司，我们限制了股票流通，花更多时间与他们沟通，帮助他们理解这意味什么。

With other companies, we restricted the stock, we spent more time with them, we helped them understand what this means.

我们甚至帮助他们中的一些人思考如何向投资者传达他们在这一转型路径上的计划。

We even helped a bunch of them think about how to communicate what they were gonna do in this path of this transition to their investors.

对我来说，这与2025年我们所处的情况不同，因为当时我们确实具备这方面的专业能力。

That to me is different than where we are in 2025, because I felt we had real expertise there.

而且我们还能为对话增添一些这些高管作为从业者此前从未见过的见解。

And something to add to the conversation that a lot of these executives hadn't seen before as operators.

坦率地说，他们的许多董事会成员都是在资金唾手可得的时代成长起来的。

And frankly, many of their board members had come of age in a period of time where money was free.

坦率地说，他们可能没有像我们那样深入参与过这么多具有持久性的公司。

And frankly, probably hadn't been involved as many durable companies as we had been involved in.

所以答案是，我认为我们更多地回到了学习和正常互动的状态，而不是急于向别人解释我们的某种观点。

So the answer is, I think probably we're more back to learning and normal interaction than we are having a perspective that we're dying to explain to people.

我在他们的早期T阶段听到了这一点。

I heard that in their early T.

Rowe时代，你研究媒体，把二十到三十年的媒体历史浓缩成一份只有三到四页的报告。

Rowe days that you were studying media and you studied twenty or thirty years worth of media history and condensed it down to a very small three or four page report.

你能回顾一下那项研究，以及你从媒体中学到了什么吗？

Can you bring us back to that study and what you learned about media?

我显然对媒体感兴趣。

I'm obviously interested in media.

我想知道你当时对媒体有什么样的认识，以及这些认识自那以后是如何演变的。

I'm curious what you learned then about media and how that has evolved ever since.

我倾向于这么做。

I tend to wanna do this.

我始终相信，如果你真正理解了某件事，就能把它说得极其简洁。

I always believe that if you really understand something, you can make it super concise.

在机器人技术方面，我们在AI领域则不然，我们的内部备忘录可能太长了，因为坦白说，还有很多未知因素，所以我们无法做到简洁。

And where we are with robotics and less so with AI, our internal memos are probably too long because frankly, there's too much unknown, and so we can't be concise.

我在媒体领域非常幸运，这也是我们在Durable公司试图对员工做到的事情。

I was very lucky in media, and it's something we try to do at Durable with people.

因为我对媒体行业来说是个外行，我认为我为它带来了全新的视角。

Because I was an outsider to the media industry, I think I brought fresh perspective to it.

当我被任命为媒体分析师时，这听起来很难相信，但当时被视为在竞争力护城河和强劲增长之间取得平衡的明星公司，是康卡斯特、时代华纳、迪士尼、维亚康姆这样的企业。

When I was assigned to be a media analyst, this is so hard to believe, but the companies that were viewed as the darlings of balancing durability in terms of competitive moats and having strong growth were companies like Comcast, Time Warner, Disney, Viacom.

我对这些公司逐一做了大量研究，然后开始深入思考这个问题。

I did a bunch of work on the companies individually, and then I started to really think hard about it.

我逐渐意识到，它们当中最好的业务其实是有线电视网络。

And I started to realize that the best businesses inside of all of them had been the cable networks.

如果当时你阅读关于媒体的资料，最著名的创业者就是那些创办了有线电视网络的人，比如约翰·亨德里克斯、泰德·特纳和鲍勃·约翰逊。

If you read about media back then, the entrepreneurs that became the most famous were the ones that launched cable networks, John Hendrix, Ted Turner, Bob Johnson.

顺便说一下，约翰·马龙几乎支持了所有这些人，并为他们提供了最多的资金。

By the way, John Malone basically backed almost all these people and put the most in business.

所以约翰·马龙处于这一切的核心位置。

So John Malone was at the center of all this.

二十年来，有线电视网络以20%的年增长率和20%的净资产收益率持续增长，因此它们是复合型增长企业。

For twenty years, cable networks grew 20% with 20 ROEs, so they were compounders.

正因如此，才出现了这么多成为亿万富翁的创业者。

And that's how you had all these entrepreneurs that had become billionaires.

也正是因此，媒体公司才会激烈争夺内容与分发之间的平衡，以便都能公平地分享这些经济收益。

And that's how you had media companies that really fought over a balance of content and distribution so they could all get their fair share of these economics.

当你去观察其他许多行业时，会发现它们的资产回报率只是平均水平。

And then when you went and you looked at a bunch of the other industry, it was like an average ROA business.

而整个行业原本就是这样的。

And that was what the whole industry was.

但如果你从系统层面认真思考，就会发现这一切都建立在一个封闭的体系之上，这一点如今显而易见。

But the whole thing, if you really thought about it at a systematic level, was predicated on a closed system, which is so obvious today.

封闭系统的基础是，我会在你最想观看、而我能赚取最多利润的时候，只向你推送你最想看的节目或电视剧。

Well, the closed system was predicated on, I'm only gonna show you the product or the TV show that you most wanna watch when I can make the most money, when you wanna consume it.

即使在有线电视时代，尽管最多人观看电视的时间是周日晚上，但最差的节目却安排在周日晚上，因为人们在周末花光了钱，又要回去工作，不会再出去购物或娱乐。

Even in a world of linear TV, even though the most people watched TV on Sunday night, the worst shows showed up on Sunday night because people had spent their money on the weekend and they were going back to work, and they weren't gonna go shopping and go out.

而最好的节目则安排在周四晚上，比如《老友记》和《考斯比一家》。

And the best shows showed up on Thursday night, Friends, The Cosby Show.

在电影行业，显然存在窗口期策略。

And in the movie business, obviously, there was windowing.

我当时就想，好吧。

I was like, okay.

所以最好的生意就是有线电视频道。

So the best business is cable networks.

这就是为什么我们会在内容与分发之间产生如此激烈的争夺。

That's why we have this fight over content distribution.

而这一切都建立在这个封闭系统之上。

And then it's all predicated on this closed system.

我当时相信的观点，显然后来被证明是正确的，那就是这场曾让众多投资者血本无归、当时没人认为能从中产生任何积极成果的TMT泡沫，实际上为该行业持久性的终结埋下了种子。

What I believed at the time, which obviously proved to be true, was that this TMT bubble that had burned so many investors and no one wanted to think anything good could come out of at the time, it laid the seeds of the end of the durability of that industry.

因为尽管人们在电信基础设施和宽带建设上损失了巨额资金。

Because even though people lost so much money on telecom infrastructure and laying the seeds of broadband.

但宽带最终促成的正是YouTube和Netflix这样的平台，它们彻底打破了整个这套体系。

What broadband was enabling eventually was things like YouTube and Netflix, which would break down this whole comp.

这个像寡头垄断一样运作的封闭系统。

This closed system that was run like an oligopoly.

我的备忘录总结指出，最危险的事就是持有那些持久性资产。

That's what my memo summarized was the riskiest thing is to own the durable asset.

而最安全的做法是去投资下一个标准。

And the safest thing to do is go buy the next standard.

我的意思是，很多人会说投资是一项学徒制的工作。

I mean, lots of people will say investing is an apprenticeship business.

而你自己也说过，最好的投资者在70岁时比50岁时更出色，而50岁时又比30岁时更出色。

And you yourself have said the best investors are better at 70 than at 50 than at 30.

我很想听听你关于如何在不太了解一个人的情况下挑选优秀人才，以及如何随着时间推移帮助他们成长的经验。

I would love to hear a lot about what you've learned about selecting great people when you don't know them as well, and then making them better as part of over time?

因为这将决定你作为企业的表现，所以这是个关键因素，你是怎么做到的？

Because that's gonna determine how well you do as a business, so it's a critical component, how do you do it?

当我们创立Durable时，我最重要的目标之一就是打造一家在我和初始合伙人离开后，比我们还在时表现更出色的投行。

One of the major goals I had when we started Durable was to actually build an investment firm that would be better the day I left and the initial partners left, than when we were the best while we were there.

所以我对此进行了深入思考。

So I thought really hard about that.

这是一个非常艰巨的目标。

It's a hell of a goal.

我开展了一次倾听之旅，去拜访那些曾经辉煌过的公司。

I went on a listening tour and I went to see firms that had a period of greatness.

其中一些公司并没有实现这一目标。

And some of them didn't get it done.

而另一些公司则确实达成了这个目标。

And then some of them actually accomplished that goal.

我从中学到了很多。

I learned a lot in that.

其中一部分是我们如何设计Durable的激励机制以及我们内部的整体理念。

Part of it is how we have structured durables incentives and the whole ethos we have internally.

但这真正强化了关于人才的这一目标。

But it was really a reinforcement of this goal about people.

你花时间与我们的团队相处。

You spend time with our team.

当你观察Durable的资深人员时，我的合伙人之一阿努克·戴，是一位非常有才华的女性，她26岁时就开始和我一起工作。

When you look at the senior people at Durable, Anuk Day, one of my partners, incredibly talented woman, she started working with me at 26.

她之前从未从事过投资行业。

She had never worked in the investment business before.

她刚从牛津大学完成硕士学位，当时在一家非营利组织工作。

She had finished her master's at Oxford and she was working at a nonprofit.

科里·肖尔，他要么在2021年、要么在2022年刚从威廉与玛丽学院毕业就加入了我。

Corey Scholl, he started working with me either '21 or '22 right out of William And Mary.

我们还有一大批来自文科背景、非顶尖学术背景的人。

And then we have a host of other people who came out of liberal arts, not rigorous.

我曾经在银行和投资公司工作过。

I had to work at a bank and an investment firm.

每当我面试别人时，我都得告诉他们，从五岁起我就想做这个了——这几乎是现在必须说的话。

Or anytime I interview with people, I have to tell people since I was five, which is basically what you have to do nowadays.

去大多数投资公司或银行面试时，你都得说，从五岁起我就想做你现在做的这份工作。

Go work at most investment firms or banks, you gotta tell people that ever since the age of five, I wanted to do exactly what you're doing.

但说正经的，我们真的相信，你必须成为自己领域的专家，并逐步成长。

But all jokes aside, we really believe that you have to be an expert in what you do, develop into it.

我们有一整套矩阵，用来衡量证券分析能力的提升过程，以及这个旅程是怎样的。

There's a whole matrix we have about the development of security analysis excellence and how is the journey?

我们的评估就是基于这个标准进行的。

We do our reviews based on it.

我会告诉人们：看这张表，这是一段旅程。

I tell people, look, on this sheet, this is a journey.

我可能是公司里最有经验的证券分析师，但我依然还有很长的路要走。

I'm probably the most experienced security analyst at the firm, and I still have a journey to go here.

我得变得更好。

I gotta get better.

同时，我们也相信，投资团队中最年轻的人往往能提供最有价值的见解。

And then we also believe at the same time that the youngest person in the room on our investment team actually can have the most valuable perspective.

我们的投资团队共有12人。

So we have an investment team of 12 people.

在我的职业生涯中，很多时候最好的洞察都来自那些以全新视角看待问题的初级员工。

And in my career, lot of times the best insight comes from the most junior person who's looking at something with a fresh set of eyes.

在阿努克职业生涯早期，当我研究消费类公司时，她帮助我深入理解了千禧一代的心态，之后我们围绕这一点做了大量工作。

Early in Anuk's career, when I was looking at consumer companies, she really helped understand a millennial mindset, and then we did a lot of work against it.

我认为这促成了我们在公开和私募市场中的一些出色投资。

And I think that led to some great investments both in the public private markets.

我们曾是DoorDash、Sweetgreen、Corby Parker等当时领先企业的私募投资者。

We were private investors in DoorDash, Sweetgreen, Corby Parker, some of the leading companies of the day.

总之，我们寻找什么样的人？

Anyway, what do we look for?

我们寻找深刻的好奇心。

We look for deep intellectual curiosity.

如果你不想持续学习，那你就不是我们这样的人。

If you don't wanna constantly learn, that's just not who we are.

我们想要真正渴望学习的人。

We want people who really wanna learn.

我们想要那些愿意竞争，但把竞争当作团队运动的人。

We want people who compete, but wanna compete as a team sport.

我们团队里有很多运动员，也有很多靠打工赚学费的人。

We have a lot of athletes, we have a lot of people who work their way through school financially.

我们想要有韧性的人。

We want people who are resilient ourselves.

我们每个人都会经历犯错的时期。

All of us have periods of time where we get things wrong.

而在市场波动如此剧烈的环境下，即使投资像科勒斯这样优秀的公司，你也必须接受有时你的表现并不会好的事实。

And then if you're gonna pass this our style of investing in a world where the market has such volatility, even on good companies like Colliers, you have to live with the fact that sometimes your performance isn't gonna be good.

有时你必须保持韧性，并意识到自己是对的。

Sometimes you gotta be resilient and you realize you're right.

你必须相信它。

You gotta believe in it.

有时你必须承认自己错了。

And sometimes you gotta realize you're wrong.

所有这些显然都建立在追求卓越并让同事变得更好的强烈愿望之上。

All this is underlied obviously by a level of desire to be excellent in what you do, but make your colleagues better.

这在杜拉尔是独一无二的。

This is unique to durable.

我们在这里就是不一样。

And we're just different here.

所以当人们思考在杜拉尔追求卓越时，他们必须思考如何在自己的工作中做到卓越。

So when people think about being excellent at durable, they have to think about being excellent in what they do.

他们还必须擅长帮助同事变得更好。

And they have to be excellent at making their colleagues better.

这两者都必须具备。

It's gotta be both.

我们是一个追求兼得的文化。

We're an and culture.

然后，显然我们谈到了我们所做的事为何与众不同。

Then obviously we talked about why what we do is different.

我们有能力投资像Duolingo这样的私营公司，并且至今仍持有它的股份。

Our ability to invest in Duolingo as a private company and still own it today.

我们有能力在Figma还是3000万美元估值的公司时进行投资，而同一个人在一天之内就实现了蜕变。

Our ability to invest in Figma when it's a $30,000,000 company and the same person evolves a day.

我们必须拥有这样的人才：既能出色地分析处于早期成长阶段的私营公司，又能理解规模化、持久增长型公司，懂得私募市场的细微差别、人际关系和治理方式，同时也真正理解作为公开市场的少数股东意味着什么。

We have to have people who are both good at analyzing private companies that are early stage growth companies and then understand scale durable growth companies and understand the subtleties and the nuances of private markets and the relationships and the way governance work, but also understand truly being a minority investor in the public markets.

第二点，即你被期望帮助同事变得更好，这在实际中是如何运作的？

That second piece which is you're expected to make your colleagues better, how does that actually work in practice?

人们具体做些什么？

What do people literally do?

这是那种模糊的、看到才懂的东西，还是更有结构性一些？

Is it squishy, no when you see it type stuff or is it more structured than that?

听好了，我先告诉你衡量标准，再告诉你实际是怎么运作的。

Look, I'll give you the measurement and then I'll give you how it really works.

在Durable进行360度评估时，我们会要求每个人提供反馈，说明他们的同事做了什么来帮助自己进步。

When we do three sixty reviews at Durable, we actually ask everyone to give feedback on what their colleagues did to make it better.

这很重要。

It's important.

我们的文化非常融洽，而且人才素质很高。

And we're a really pleasant culture and we have high quality people.

但这不是‘Anuk今年帮了我，她人很好’这样的说法。

But it's not Anuk helped me this year, and she's a nice person.

而是如果你关注DoorDash，那就是‘Anuk帮助我理解了DoorDash，因为她对Shopify上代理商业务的深刻见解真的帮到了我’。

It's if you're following DoorDash, it's that Anuk helped me understand DoorDash because of her knowledge about agent commerce on Shopify really helped me.

或者当我们作为投资团队进行投资评审时，她特别关注并主动跟进我，或者她参加了重要的会议，并向我分享了她的见解。

Or when we did an investment review as an investment team, she took a special interest and she followed up with me, or she went to a meeting that was important and gave me her perspective.

我们要求人们具体指出他们所参与的投资项目。

We ask people to basically point to specific investments that they have.

第二点，我是那种相信要想拥有优秀人才，就必须吸引优秀人才，让优秀人才成长，并提供相应环境的人。

The second thing we do is, I'm one of these people who believe we wanna have great people, we have to attract great people, we have to allow great people to become great and provide the environment.

但我同时也相信，我们必须有恰到好处的流程，以支持真正的卓越与创造力。

But I also believe we have to have the right amount of process that enables true excellence and creativity.

从时间投入的角度来看，我认为我们做的比其他投资公司少，或者至少和他们一样多，但我觉得我们的影响非常大。

We, I think from an hour perspective, probably do less or at the same amount as other investment firms, but I think the impact is really high.

我给你举个简单的例子。

I'll give you a simple example.

我们和其他公司一样，每周一都开同样的投资会议，但我们会额外增加一场会议，一起讨论我们正在关注的项目，并共同筛选。

We do the same investment meeting everyone else does on Mondays, but we do an additional meeting where we go through ideas that we're looking at and we gate them together.

我们会花更多时间在某个投资项目上。

We're gonna go spend more time on an investment.

当我们决定要去做某件事时，每个人都在场。

Everyone's in the room when we decide to go do it.

所以人们开始明白，哪些是他们时间的合理用法，哪些不是。

So people start to learn what is it that's a good use of their time, what's not a good use of a time.

如果是的话，如果我们打算去研究某个项目，而我在下一阶段的尽职调查中需要回答几个问题，这里有没有同事能帮我？

And if it is, if we're gonna go look at something and I gotta go answer these couple questions in the next stage of investment due diligence, is there a colleague here who can help me do it?

我们在周五聚在一起。

We get together on Fridays.

顺便说一下，我们在办公室进行。

And by the way, we do it in the office.

我们和投资团队一起吃午餐，讨论各种见解。

We have lunches and investment team, and we talk about insights.

你不需要为这个会议做准备，但你可以分享：我最近和一位CEO进行了一次有趣的对话，或者这周我们肯定会谈到，我听了OpenAI开发者大会，我觉得这些内容很有趣。

You don't prepare for this meeting, but this would be, hey, I had an interesting conversation with a CEO, or maybe this week I'm sure we'll talk about it, I listened to OpenAI Dev Day, and this is what I thought was interesting.

或者我思考了这个问题，有人也这么想吗？

Or maybe I thought about this, does anyone think about this?

所以我们试图寻找横向的洞察，互相学习。

So we're trying to look for lateral insights to learn from each other.

当然，我们还会进行投资评审，深入分析个股。

And then for sure, we do investment reviews where we do deep dive on stocks.

有多个人会一起研究这些股票。

Multiple people look at this.

我从你提到的凯利在Lone Pine公司的情况中学到了这一点。

I learned this from you had Kelly on the firm Lone Pine, really.

我是从史蒂夫那里学到的。

I I learned that from Steve.

我肯定Lone Pine现在还在这么做。

I'm sure Lone Pine still does it.

我们会做关键绩效指标（KPI）。

We do KPIs.

我们把KPI称为运营复盘，会全面回顾整个投资组合，每位同事都会向其他人汇报自己的表现。

We call KPIs as operating reviews where we go through the whole portfolio, and every colleague reports to everyone else how they did.

我们这样做并不是为了搞成一个只是夸赞‘你太棒了’或者‘我搞错了’的会议。

And we do this not to basically be a session, wow, you're great, or I got it wrong.

更准确地说，是客观地呈现发生了什么，每个人都能从中学习，同时也能贡献自己的视角。

It's more like, here's clinically what happened, and everyone can learn from it, but also lend their lens.

因为这并不是非好即坏。

Because it's not great, bad.

很多时候，情况是很微妙的。

A lot of times it's subtle.

这件事做得不错，那件事则好坏参半，人们无法从自己的视角去理解。

This thing was good, this thing was mixed, people couldn't have their lens.

然后我们每年会聚两次，举行外出研讨会。

And then we get together twice a year and we have off sites.

你大概能感觉到，我们既有趣又持久，但我们的外出研讨会和其他人的完全不同。

You can probably tell we have a lot of fun and durable, but our off sites don't look like other people's off-site.

我们以前会做团队建设活动。

We used to do team building activities.

我负责KPI

I do KPI

评审。

reviews.

现在我们不做这些了。

And now we don't do them.

我们为什么不做了？

Why don't we do them?

因为我们其实是一群喜欢互相学习、分享见解的人。

Because actually we're a group of people who likes learning from each other and like sharing insights.

我们为期三天的活动将主要是进行回顾，查看评审内容，研究行业，与首席执行官交流，然后大家从中学到东西，从而互相提升。

And the activity where we got together for three days is gonna be we're gonna basically do look backs, we're gonna look at reviews, we're gonna go study an industry, we're gonna go talk to a CEO, and then we're gonna all learn from something such that we can all make each other better.

当你对KPI或其他指标做出初步预测时，知道在六个月、一年、两年甚至三年后会被回溯审视，这一定非常有力量。

It must be very powerful when you're making an initial projection on a KPI or something to know that in the six months and year and two years and three years hence, it's gonna be looked back upon.

你可能会更仔细地推敲。

You probably sharpen your pencil a little bit.

你会的，但我认为我们文化特别之处在于，我总是告诉即将加入的人，我们倾向于招聘年轻人并培养他们。

You do but I think what's special about our culture, and I always tell this to people before they join, we tend to hire young people and develop them.

但如果你曾在其他地方工作过，你不会相信我们。

But if you've been anywhere else, you don't believe us.

人们认为，当我们聚在一起做KPI或三年回顾时，有时我们会分析投资组合中的重新启动项目——那些我们之前出售过又买回来的东西。

People think when we get together and we do KPIs or do three year look backs, or we'll do a session sometimes where we'll look at reinitiations in the portfolio, things we sold before and then we bought back.

很多时候，如果我们这么做，为什么会一开始卖掉呢？我们是不是判断错了？

And a lot of times if we did that, why did we sell it in the first place and we got it wrong?

我们从不以‘你犯了错’的心态来做这些事，这里不是一个让你自我否定批评的环境。

We don't do any of this in the spirit of you made a mistake, this is an environment where you should critique yourself negatively.

相反，我们应该保持智力上的诚实，就像我们希望我们的高管那样。

It's like, no, we should be intellectually honest, just like we want our executives to be.

我们希望对所做的事情保持客观冷静。

We want to be clinical in what we did.

但同时，我们希望以学习和进步的心态来做这些事，或者把我们的数据公开出来，以便从可能拥有宝贵见解的同事那里学到东西，让我们变得更好。

But then we want to do it in the spirit to try to learn from it and get better, or put our data out there so we can learn from our colleagues who might have a valuable perspective to make us better.

如果我们能秉持一种理智、诚实、自我提升和谦逊的态度，这会让我们变得更好。

And if we can take the attitude of intellectual, honest, self improvement, humility, that like that makes us better.

当你在走访了解那些在创始人任期结束时表现优异的业务单元，以及那些未能成功的业务时，你学到了什么？

When you were doing your tour to learn about the franchises that were better at the end of the founders run and those that didn't make it, what did you learn?

我不会提及那些未能成功的公司的名字。

I won't mention the names of those who didn't make it.

因为未能成功这件事，有点像我们投资时写的备忘录——当我们投资那些努力创业的优秀人才时。

Because the thing about not making it is a little bit like our investment memos when we invest in great people who are trying to build companies.

他们中的许多人做了很多了不起的事，但最终还是没能成功。

A lot of them do great things and they just don't make it.

成功往往包含很多好运。

In success, there's a lot of good fortune.

我想我学到的是，如果你从第一天起就没有为成功设计好体系，那么最终会陷入大量冲突，有时这些冲突会破坏你本可以实现的成就。

I think what I learned was if you don't architect the system on day one for success, then you end up with a lot of conflicts that sometimes undermine what you could have accomplished.

我们在Durable公司竭尽全力避免做出妥协。

We try really, really hard at durable not to make compromises.