本集简介
双语字幕
仅展示文本字幕，不包含中文音频；想边听边看，请使用 Bayt 播客 App。
欢迎收听EconTalk，这是经济学与自由图书馆旗下的节目。
Welcome to EconTalk, part of the Library of Economics and Liberty.
我是您的主持人，来自乔治·梅森大学和斯坦福大学胡佛研究所的罗斯·罗伯茨。
I'm your host, Russ Roberts of George Mason University and Stanford University's Hoover Institution.
我们的网站是econtalk.org，您可以在那里订阅节目、收听往期内容、留言评论，还能找到与今天对话相关的链接和其他资料。
Our website is econtalk.org, where you can subscribe, find other episodes, comment on podcast, and find links and other information related to today's conversation.
我们的邮箱是mail@econTalk.org。
Our email address is mail@econTalk.org.
我们非常期待听到您的声音。
We'd love to hear from you.
今天是7月17日，我的嘉宾是保罗·格雷厄姆，作家、程序员，也是编程语言的设计者。
Today is July 17, and my guest is Paul Graham, essayist programmer and programming language designer.
他著有《黑客与画家》等多部作品，同时也是Y Combinator的合伙人。
He's the author of among other books, hackers and painters and a partner in the y combinator.
保罗，欢迎来到EconTalk。
Paul, welcome to EconTalk.
哦，谢谢。
Oh, thank you.
咱们先聊聊 Y Combinator 吧。
Let's start with the y combinator.
你给我们讲讲它到底是怎么回事。
Tell us what that is.
它是个风险投资公司，但跟一般的有什么不一样？
It's a venture capital firm, but what's different about it?
你当初为什么会参与进来？
Why'd you get involved in it?
我们其实不太愿意把它叫做风险投资公司。
We wouldn't we wouldn't use the term venture capital firm to describe it.
因为在业内，'风险投资公司'这个词有非常特定的含义。
Because in the business, venture capital firm means something very specific.
那通常是指你从大学基金会之类的地方筹了上亿美元，然后有一堆投资合伙人坐董事会什么的。
It means, like, you raised a couple $100,000,000 from university endowments or something like that, and then, you know, you have a bunch of, like, venture partners who sit on boards and stuff like that.
我们做的是种子轮投资。
What we do is seed funding.
对。
Yeah.
目前我们还没有一个确切的名称，因为我们做的事还算新颖。
And there's not really any name for us yet because what we do is sort of new.
我其实挺高兴的。
And I'm actually glad.
我一直主张别给我们定名，因为这样一来，但凡有人想模仿我们，就只能直接提到我们。
I have I have encouraged there to be no name for us because what that means is whenever whenever anybody starts some kind of new thing copying us, the only way they can describe it is to mention us.
没错。
That's true.
他们只能说，像Y Combinator那样的模式。
They have to say, y-combinator like thing.
对吧？
Right?
当然。
Sure.
但如果有个通用词来描述它，他们就可以直接说‘一个新的空白’。
Whereas if there were a general noun for it, they could just say a new blank.
没错。
That's true.
对吧？
Right?
所以我挺高兴我们没什么特定的名称。
So I'm happy that there's no name for what we are.
那你们到底做些什么？
So tell us what it does.
对。
Right.
所以，我们冒着可能给别人提供机会的风险，去创造
So we At the risk of creating a opportunity for somebody to then create
一个名字
a name
叫它。
for it.
但不妨试试看。
But take a chance.
是的。
Yeah.
他们一直想用‘孵化器’这个词，但这个词已经有别的意思了。
They've been trying to use the name incubator, but that already meant something else quite different.
明白了。
Okay.
我们做的和风险投资基金最大的不同是，风险投资基金只下少数几个大注。
What we do is, the big difference between us and a venture capital fund is that a venture capital fund makes a small number of large bets.
对吧？
Right?
风险投资基金每次投资大约一百万到两百万美元，每个合伙人每年大概只做两笔投资。
A venture capital fund will invest on the order of 1 or 2,000,000 at a time, and they will do maybe two deals per partner per year.
我们只投两万美元，但四个合伙人一年就能投进五六十家初创公司。
We invest $20,000, and we might invest in 50 or 60 startups in a year with just four partners.
对吧？
Right?
相当于每个合伙人做十五笔投资。
So 15 deals per partner.
交易量是七倍，但单笔金额只有百分之一。
Seven times the volume and one hundredth the amount.
那成功率大概是多少呢？
And, what's what's the likely, success rate?
说实话，我们现在还不知道。
You know, we don't know yet.
这跟其他模式相比还不太好说。
It's relative to other model.
不过成功率不会那么高。
Well, it's not gonna be as high.
事实上，如果我们的成功率跟顶级风投一样高，我反而会担心，那说明我们太保守了。
In fact, if our success rate was as high as a good VC funds, that would that would worry me because it would mean we were being too conservative.
对吧？
Right?
要是失败的项目不多，那我们就太谨慎了。
This this this There ought to be a lot of failures or we're being too careful.
虽然我现在还不知道成功率会是多少，但如果能达到三分之一，我会很满意。
So although I don't know yet what the success rate will be, I would be happy if it were a third.
对吧？
Right?
这意味着我们投资的初创公司大多数都会失败。
Which means most of the startups we fund will fail.
但这没关系。
And that's okay.
这很正常。
That's standard.
对。
Yeah.
是的。
Yeah.
所以你们投的金额不大，是怎么筛选项目的？有没有什么规律可循？比如公司类型或者创意方向？
And so you're giving a small amount, and how do you screen them, and and is there any pattern to what kind of firms there kind of ideas there?
确实有规律，但不是在创意上。
There are patterns, but not not in ideas.
我们是批量处理的，就是计算机里说的批量模式。
We screen them We we do everything in batch, what what computer people call batch mode.
如果要用一句话概括我们的做法，那就是把流水线生产的方式用在风险投资上。
What we're doing, if you wanna if you wanna understand what we're doing in one sentence, it is we are applying mass production techniques to venture funding.
对吧？
Right?
其实我们一开始并没打算这么做。
Now, we didn't set out to be doing this.
大概六个月后我们才意识到，原来我们一直在把大规模生产的方式用在风险投资上。
We realized about six months in, you know, what we've been doing is we've been applying mass production to venture funding.
对吧？
Right?
在经济史上，你总能反复看到这样的模式：某样东西原本是手工一件一件做，又贵又不稳定。
There's this pattern you see over and over again in economic history where there's something that's made one at a time very expensively and unreliably.
对吧？
Right?
然后有个人出现，想办法优化整个流程，用标准化的方法一次性大批量生产出来。
And someone comes along and figures out how to, like, rationalize the process and produce a whole bunch of them all at once using standardized procedures and stuff like that.
对吧？
Right?
通常是通过在流程中引入大量新技术，或者不同的技术来实现的。
Usually by adding a lot of technology to the process or a different kind of technology.
是的。
Yes.
我们就是这么做的。
And that is what we do.
我们就是这么做的。
Is what we do.
我们一次性投资一大批初创公司，把它们都带到门洛帕克来。
We fund a whole bunch of startups at once, and we bring them here to, you know, Mountain View.
它们全都同时启动。
And they all start at the same time.
不是30家。
30 no.
26家初创公司全都
26 startups all
真的是同时启动吗？
start Literally?
是的。
Yes.
就像一个集训营。
Like a camp.
你可以说，这是出发点，或者像一场赛马。
You sort of say, here's out or maybe a horse race.
开赛那天，就是起跑线。
Opening day, here's the starting line.
那不如说是个研究生项目？
How about a graduate program?
对。
Okay.
因为这其实最像一个研究生项目。
Because that's what it's really most like.
真的就像某种研究生项目。
It's really like some kind of graduate program.
如果你非得找一个现实中已有的东西来比喻，那它更像研究生项目，而不是新兵训练营。
If you had to find something else that already existed in the world to stick it to you, it's more like that than a boot camp.
也许更多人去过新兵训练营，没怎么接触过研究生项目，所以他们更喜欢用这个比喻。
Maybe more people have been to boot camps than graduate programs, and so they prefer that metaphor.
对吧？
Right?
不对。
No.
但它根本不像新兵训练营。
But it's it's not like boot camp.
我们不会让人做俯卧撑。
We don't make people do push ups.
对吧？
Right?
我们也没有那种吓人的教官。
We don't have fearsome drill instructors.
你们会教一些内容吗？
Do you teach some stuff?
会的。
Yes.
哦，
Oh,
对。
yes.
那普遍的主题或者大规模生产部分是什么呢？
So what's the what's the common theme that, or the common materials that's the mass production part?
你记得你说过，这总是涉及技术的引入。
Well, you remember you said it always involves introduction of technology.
对吧？
Right?
每当我能为 Y Combinator 写软件时，我都特别开心，因为这意味着我能把原来手动做的事变成程序，一敲代码，事情就自动完成了。
Whenever whenever I can write software for y combinator, I'm always happy because that means I can take something that we were doing manually and I can embody it in software and bang bang bang, it just happens.
对吧？
Right?
所以我们内部用了很多自研的软件。
And so we have a bunch of software internally that we use.
你之前问过我们怎么筛选申请人。
One of the things you asked was how we screen people.
每期我们都会收到几百份申请。
We get many hundreds of of applications for every every batch.
我们一年办两期，一期在六月，一期在一月。
We do two batches a year, one starting in June and one in January.
对吧？
Right?
每次申请都成百上千，这次可能快到一千份了，因为上一次就快到一千了。
And every time we get, like, hundreds and hundreds of applications, we might actually get a thousand for this next batch because we're getting close to a thousand last time.
要是靠纸质表格，或者得一个个看PPT，我们根本看不完一千份PPT。
And like, if we did this, you know, with printed forms or something like that or if we had to read if we had to look at people's PowerPoint decks, we couldn't look at a thousand PowerPoint decks.
我们根本做不到。
We just couldn't do it.
对吧？
Right?
但那样的话，你就得雇很多人，还得盯着他们，确保他们给的建议靠谱。
So But it would you'd have to add a lot of people, then you'd have to monitor the people, make sure they were giving you good advice.
没错。
Right.
没错。
Right.
我的意思是，我们
Mean, we
我的意思是，我和Y Combinator的其他合伙人，总共就四个人。
I mean, me, the y-combinator partners, there's only four of us.
对吧？
Right?
这些材料我们都得自己看。
We have to read all these things ourselves.
而且我们还有别的事要做，比如罗伯特是麻省理工的教授。
And we have other things to do, you know, like Robert's a professor at MIT.
所以我们现在用的是申请表。
So what we do is we have an application form.
我们不会去看任何人做的PPT。
We're not gonna look at anybody's PowerPoint presentations.
我们用的是申请表。
We have an application form.
每个人都回答同样的问题。
Everybody fills out the same questions.
然后我们用一套在线软件快速浏览这些申请，筛选、打分，挑出表现最好的那些。
And then we have software online that we we use to blow through these applications and read them quickly and rank them and grade them and look at the top ones.
申请表做得最好的，我们会邀请他们来面谈。
The ones that do best in the application form, we invite to meet us in person.
下一波申请者我们会安排面试。
We have interviews in, this next batch.
面试定在十一月的那个周末。
The interviews will be the weekend of November.
对吧？
Right?
一个周末我们可能会面谈六十家初创公司，每天二十家。
And we will talk to maybe 60 startups in one weekend, 20 a day.
对吧？
Right?
每家我们只谈十分钟，十分钟后就能决定要不要投资，因为这就是必须做的。
We talk to them for ten minutes each and we decide whether to fund them or not after ten minutes because that's what you have to do.
没错。
Right.
就像在摩天大楼里做电梯推销一样。
Elevator pitch with a long a tall building.
其实也不是。
Well, no.
我们其实不太喜欢别人向我们推销。
Actually, we don't like it when people pitch us.
我们没时间听人推销。
We don't have time to be pitched.
对吧？
Right?
我们通常是问他们问题。
What we do is we ask them questions.
对。
Right.
就是问他们问题。
We just ask them questions.
他们一进来坐下，我们就开始提问。
They walk in and sit down and we start asking questions.
我们就问他们十分钟的问题。
We just ask questions for ten minutes.
每次面试你们四个人都参与吗？
Do all four of you participate in each interview?
是的。
Yes.
对。
Yes.
他们进来的时候，感觉就像一场军事法庭审讯。
They come in and it's sort of like a court martial.
没错。
Yeah.
我们四个人坐在那儿，他们进来坐下。
It's four of us sitting there and they come in and sit down.
我们其实会试着让他们别紧张，但这种场合难免让他们紧张，你也知道。
And we try and encourage them not to be nervous actually, but I think the situation inevitably makes them nervous, but, you know
这简直像在读研究生。
That's like graduate school.
是的。
Yeah.
我真不喜欢让别人紧张，说实话。
I hate it that it makes people nervous actually.
不只是因为我同情他们，而是因为紧张会让面试变得困难得多。
Not just because I sympathize with them, but because it makes the interview so much harder.
而且也更没意思，
And less fun, by
顺便说一下。
the way.
对。
Yeah.
我们本来想问他们一些问题，可他们却连话都说不利索。
We're trying to ask these guys questions and they're like stumbling over their words.
对吧？
Right?
我们只是想从他们那里把信息问出来。
We just wanna get the information out of them.
如果他们紧张，情况就会更糟。
If they're nervous, it makes it go worse.
是的。
Yeah.
所以我们尽量让每个人放松，态度友好，但他们还是难免紧张。
So we try as much as we can to calm everybody down and be very friendly, but they're, you know, they're nervous regardless.
你做这个多久了？
How long have been doing it?
四年了。
Four years.
我们是从2005年开始的。
We started the 2005.
今年夏天这一批是第九批初创公司了。
So this summer's batch is the ninth batch of startups.
那第一批现在怎么样？
And how's the first batch doing?
第一批实际上表现得特别出色。
The first batch actually did amazingly well.
说实话，可能好得有点不正常。
In fact, probably did anomalously well.
我们大概就是运气特别好。
We probably just completely lucked out.
对吧？
Right?
但第一批里有Looped这家公司。
But the first batch included, Looped.
不知道你有没有听说过他们。
I don't know if you've heard of them.
他们可能在另一个年龄段更受欢迎。
They're probably mostly popular in a different demographic.
这是一个手机应用，能显示你朋友当前的位置地图。
They're this application that shows you on your cell phone a map showing where all your friends are.
是的。
Uh-huh.
对吧？
Right?
对高中生来说，这简直是全世界最重要的事。
Like, for high school kids, this is the most important thing in the entire world.
你朋友现在人在哪儿？
Where are your friends right now?
就是实际的地理位置。
Physically, where they are.
没错。
Yeah.
对。
Right.
这些就是点。
This is the dots.
是的。
Yeah.
你能看到它们在移动什么的。
You can see them moving around and stuff.
就像手里握着一个蚁群。
Kinda like having an ant colony in your hand.
是的。
Yeah.
是的。
Yeah.
像一个蚂蚁农场。
An ant farm.
对。
Yeah.
所以 Loop 这个项目非常成功。
So that is Looped is very successful.
第一期的另一个初创公司是 Reddit。
And another startup in the first batch was Reddit.
你听说过 Reddit 吗？那个社交新闻网站。
Do you know Reddit, the social news site?
知道。
Yeah.
对吧？
Right?
它也是第一期的。
That was in the first batch.
第一期里还有 Justin.tv 的创始人。
And in the first batch, were the founders of justin.tv.
你听说过Justin.tv吗？
Do you know about justin.tv?
Justin.tv是领先的直播视频网站。
Justin.tv is the leading live streaming video site.
它是排名第一的。
It is the number one.
他们网站的流量有个特别惊人的数据，比Hulu之类的平台还多。
There's some astonishing statistic about their traffic, like they get more viewers than Hulu or something like that.
我不太确定。
I don't know.
你可以去查一下。
You can look it up.
Justin.tv的那帮人之前搞过另一个创业项目，但失败了。
The justin.tv guys were doing a different startup that failed.
不过我们挺欣赏他们，所以又投资了他们的下一个项目，结果就是Justin.tv。
But then but we liked them, and so we funded them again for their next startup, which turned out to be justin.tv.
第一批里有个创业公司，是第一个被收购的。
And there was a startup in the first batch that was the first one that got bought.
其实叫TextPayMe，但没人听说过，因为被收购得太快了。
Actually, it was called text pay me, and no one's heard of it because it got bought so quickly.
几乎是刚从那批出来就被人买走了。
It got bought, like, almost instantly out of that batch.
但我刚才打断你了。
But I interrupted you before.
你把这些人请来，面谈之后，就决定录用。
So you bring these guys in, you interview them, and then you say yes.
明白了。
Okay.
那接下来发生什么事了？
So the first Tell me what happens after that.
好的。
Okay.
好的。
Okay.
我直接快速说一下。
I'll just blow through it.
他们都在线申请。
They all apply online.
不然我得打断你了。
Or I'm gonna cut you off.
对。
Yeah.
没错。
Right.
好的。
Okay.
所以他们都在线申请。
So they apply online.
我们会邀请五六十个人来面谈。
We invite 50 or 60 to meet us in person.
整个周末累得够呛，每人只谈十分钟。
We talk to them for ten minutes apiece over one grueling weekend.
行。
K.
我跟你说啊。
Let me tell you.
那天晚上，我们就通知每个人结果，通过还是不通过。
That evening, we tell everybody yes or no.
周末结束时，我们选中的那些人会再回来，先给他们讲些基本事项，然后告诉他们，要么一月、要么六月来门洛帕克报到，看具体是哪一轮。
At the end of the weekend, the ones we said yes to come back and we'd like explain some basic stuff and then we tell them, show up here in Mountain View in either January or June depending on which one it is.
对吧？
Right?
对了，我其实还想过拍个电视节目。
And I thought about doing a TV show by the way?
你是说那种真人秀吗？
You mean like a reality show?
是的。
Yeah.
就像《美国偶像》那样，观众会特别喜欢看你们培养这些年轻人成长、学习的过程。
Like American Idols, people would love to watch you grow these guys that learn a lot.
对。
Yeah.
而且他们都很棒。
And they're cool.
机会确实不少，不过你先说。
There's a lot of opportunities maybe, but go ahead.
不好意思。
Sorry.
不用。
No.
从一开始，真人秀制作方就被要求这么做。
People have like from the beginning, people reality show producers have been asked to have like this.
但这种情况会变成这样：
But it would be one of these things where
有点破坏了原本的氛围。
Kinda ruins the
记录发生的事，反而会影响事情本身的发展。
Measuring what happened would really affect what happened.
是的。
Yeah.
不对。
No.
我明白。
I know.
这会是个失败的主意。
It'd be a loser.
不过 Anyway，你继续说。
But anyway, go ahead.
所以你把他们叫回来。
So you bring them back.
你让他们
You tell them to
回来。
come back.
关键是，我都不确定 Y Combinator 能不能行得通。
And the thing is, I don't even know if y combinator is gonna work.
对吧？
Right?
到目前为止看起来是挺有希望的，但我真不知道最终回报会怎么样。
So I don't I mean, it seems promising so far, but I don't know for sure what our returns are gonna be like.
我们目前肯定还没赚到什么钱。
We're certainly not making lots of money yet.
对吧？
Right?
如果我们正好卡在成败的边缘，我可不想做些什么把我们推到失败那边去。
And and if we're just like on the borderline of working or not working, I don't wanna mess with something that might kick us over into not working.
没错。
Right.
或者反过来也一样。
Or vice versa.
是的。
Yeah.
让你错失爆发式增长的机会。
Keep you from becoming wildly successful.
没错。
Yeah.
所以我觉得，嗯
So I'd say, So
我们刚接收了一批初创公司，大概二十五到三十家。
we've accepted a new batch of start up, say 25, 30 of them.
他们全都回家去收拾东西，然后又回到这里。
They all go back home and, like, you know, pack all our stuff or whatever and come back here.
他们大部分本来就是本地的吗？
Are most of them from here already?
不是。
No.
他们来自世界各地。
They come from all over.
全球各地都有。
All over all over the world.
上个冬季批次，十六家初创公司里有七家来自海外。
The last winter batch, seven out of 16 startups were from overseas.
对吧？
Right?
这个市场真的太大了。
It is so it is such a huge market.
是的。
Yeah.
种子轮融资这个市场。
The the the sort of seed funding market.
这是很多人不了解的一点。
That's one big thing that people don't understand.
他们不明白，种子轮融资市场并不是区域性的。
They do not understand that the seed funding market isn't regional.
他们以为自己能搞个像Y Combinator那样的项目，比如在圣路易斯之类的。
You know, they think, I'm gonna start the y-combinator of, you know, St.
然后就开始做这件事。
Louis or something like that.
可等到真去做了才发现。
And then they discover when they do it
你算是竞争对手了。
You're a competitor.
对。
Yeah.
没错。
Yeah.
毕竟大家都在争抢同一批申请人。
Well, everybody's all competing for the same pool of applicants.
对吧？
Right?
他们根本没想到——那些在圣路易斯本地的创业者们
They don't The people in St.
圣路易斯的创业者不会因为本地已经有了类似机构，就不来申请加入我们的项目。
Louis don't not apply to us just because this other thing's in St.
的。
Louis.
同时，他们收到的申请来自全国各地，不只是圣路易斯。
And meanwhile, they get applications from all over the country, not just St.
圣路易斯。
Louis.
对吧？
Right?
其实都是同一拨人。
It's all the same people.
因为在这个阶段，创始人完全是可以流动的。
Because the founders at this stage are completely mobile.
有两个带着笔记本电脑的家伙，早就习惯了睡沙发、吃外卖披萨。
There are two guys with laptops who are like used to sleeping on sofas and eating takeout pizza.
对吧？
Right?
没错。
Yep.
他们想去哪儿就去哪儿，真的会这么做。
They could go anywhere, and do.
对吧？
Right?
从经济角度看，这其实很有意思——历史上很少有群体像他们这样高度流动。
That is actually economically If you're if you're interested in the economic aspects of that, that is actually historically quite interesting that these people are so mobile.
对吧？
Right?
说得太好了。
That's perfect.
这意味着，如果某个城市对初创企业稍微友好一点，就会把其他城市的优秀人才全吸走。
It means that if a city was like a little bit better than other cities for startups, they would just vacuum all the good people out of the other cities.
是的。
Yeah.
在一定程度上是这样。
To some extent.
没错。
Right.
这事其实一直在发生，但现在比二三十年前要频繁得多。
Well, it's been happening, but now it's gonna happen much more than it did twenty or thirty years ago.
为什么呢？
Why?
因为现在创业变得容易多了。
Because it's now so much easier to start a startup.
二十多年前为什么？为什么会这样？
Twenty or Why is that?
因为创业的成本变低了。
Because it's cheaper to start startup.
明白了。
Okay.
我来跟你说说。
I can tell you.
我一直在仔细思考这个问题。
I've thought a lot about this.
现在创业成本变低，主要有四个原因。
There's four reasons it's cheaper to start a startup.
希望我能一下子把它们全想起来。
I hope I can remember them all right on the fly.
能说三个就不错了。
Three would be great.
不行，我得一个不落全说对。
Now, I gotta get them all right.
好，那我开始说。
Okay.
第一个原因是硬件成本。
So one of them is hardware.
对吧？
Right?
展开剩余字幕（还有 480 条）
摩尔定律让电脑变得几乎免费了，因为以前你要创业，还得专门去买昂贵的电脑。
Moore's law has made computers effectively free in the sense that, you know, you used to If you started to start up before, you used to have to go out and buy fancy computers to do it.
现在呢，你用来打《魔兽世界》的电脑就已经完全够用了。
And now, like, computer you've already got for playing World of Warcraft is good enough.
对吧？
Right?
所以摩尔定律让电脑的成本几乎降到了零。
So Moore's Law has made computers effectively free.
互联网让推广变得免费。
The Internet has made promotion free.
以前你做出个产品，至少得花钱打广告或者请公关公司来宣传。
It used to be if you made a product, you had to, like, buy advertising or hire a PR firm at least to get word out.
对吧？
Right?
现在信息传播快多了，不光靠口口相传，还能通过互联网上的新平台迅速扩散。
And now, word can spread so much more quickly, not just through word-of-mouth, but through new publications on the Internet.
哦，社交网络啊。
Oh, social networking.
就是说，现在找有趣的东西比以前方便多了。
Mean, just easier to find cool stuff than it used to be.
对。
Yeah.
所以，你只要做出来，别人自然就会来找你。
So, basically, you build it and they will come.
以前可没这么容易。
And that didn't used to be as true as it is now.
对吧？
Right?
这就两个了。
That's two.
明白了。
Okay.
我们现在说到第二个了。
We're up to two.
另一个变化是编程语言变得更先进了。
Another one is programming languages have gotten more advanced.
编程语言变得更抽象了，你可以用更大的积木块来搭建程序。
Programming languages have gotten more abstract in the sense that you sort of build programs out of bigger Lego blocks.
对吧？
Right?
这意味着你不用做那么多工作，就能完成同样规模的程序。
And that means you don't have to do as much work to get a given amount of program done.
在这种情况下，其实发生了一种质的变化。
And on in that case, what's happened is there's actually been a qualitative change.
过去，大家觉得创建一家初创公司，不管做什么，工作量都差不多。
Because it used to be most people, you know, to build a startup as if they were all the same amount of work.
对吧？
Right?
但假设还是那样的话，你得找五六个开发者，用 C++ 写代码。
But just assuming they are, you know, you would have to have, like, a team of developers, like, five or six people writing code in c plus plus.
对吧？
Right?
可初创公司不可能有五六个创始人。
And it doesn't work to have five or six founders in a startup.
所以很可能，你得雇人。
So probably what that means is you have to hire people.
但现在用 Ruby、Python 这类比 C++ 更抽象的语言，光靠创始人就能搞定。
Now, using a language like Ruby or Python or something like that that's more abstract than c plus plus, you can do it with just the founders.
要是两个创始人都会编程，或者三个创始人里有两个会写代码，那就完全没问题。
If you have two programmers among the founders, like you have two founders and they can both program or three founders and two of them are programmers.
对吧？
Right?
根本不用再招人了。
You don't have to hire anyone.
你自己就能搞定。
You can just do it yourself.
所有代码，第一个版本你都能自己写，这样就不需要筹钱去雇人了，因为请人真的会突然变得特别贵，你知道的。
You can write all the code, the first version yourself, which means you don't need to raise money to hire anyone because hiring people, like, that's suddenly shockingly expensive, you know.
这可是现金流的问题。
Like That for cash flow.
如果你只是创始人自己，吃泡面、在家办公，本来就已有电脑和网络，对吧？
If it's just you the founders like living on ramen and working out of their apartments and they already had computers and a network connection, right?
这比整个夏天啥也不干、闲逛的开销还低。
I mean, that's no more expensive than just like hanging out doing nothing for a summer.
对吧？
Right?
但一旦你得雇人，天啊，成本立马就飙升了。
But when you have to hire people, man, suddenly it gets much more expensive.
对吧？
Right?
所以你根本不需要雇人。
So you don't have to hire people.
说到底就是这么回事。
That's what it comes down to.
四个里有三个，但还有一个。
That's three out of four, but there's another one.
不过不管怎样，重点是成本变低了，我倒忘了我们当初为什么关心这个，不过当时是有的。
But anyway, the main point is it's gotten cheaper and I forgot why we were interested in that, but there was
这就是为什么现在初创公司多了这么多。
Well, that's why there are so many more startups.
对。
Right.
对吧？
Right?
之所以现在初创公司这么多，原因就在于
The reason there are so many more
哦，我知道。
Oh, I know.
还有他们的灵活性。
And mobility of them.
这一点也很关键。
That was what was important too.
没错。
Yes.
他们确实灵活。
Well, they're mobile.
他们灵活是因为就两个人。
They're mobile for this It's just two guys.
原因也是一样的。
Same reason that there are so many.
对吧？
Right?
现在创业成本这么低，意味着谁都能干。
When it's very cheap to start a startup, what that means is anyone can do it.
对吧？
Right?
以前创业得是搞硬件公司，得筹钱建工厂，
Like, back in the days when a startup meant, like, a hardware company and you had to get raise money to build an actual factory
没错。
Right.
那时候没几百万根本玩不转。
The only people you know, you needed millions of dollars to do this.
要拿到这么多钱，你得看起来靠谱。
And to get those millions of dollars, you had to be plausible.
得是年纪大、穿着得体，还有一长串论文发表记录。
You had to be old and well dressed and have this long list of publications.
你也许根本就是个骗子，但只要装得像那么回事，就能把钱筹到。
Maybe you could be a complete useless fraud, but as long as you looked plausible, you know, you could raise this money.
现在，任何想做的人都能做。
Now, anybody who wants to do it can.
根本不用花钱。
Doesn't cost anything.
顶多就是暑假闲着没事干而已。
Doesn't cost anything more than just hanging out for a summer anyway.
对吧？
Right?
没错。
Right.
这意味着，现在参与的人多了很多，而且多样性也高得多。
And so what that means is there's this huge, much larger pool of people and there's much more variation in it, actually.
如果回报和风险成正比，那现在大家能尝试的冒险点子比以前多得多，技术进步自然也会更快。
I mean, if, if reward is proportionate to risk, then the fact that people can do so many more risky ideas than they used to be able to automate and the technology will now progress faster.
对吧？
Right?
保罗，关注右尾部。
Paul, the right hand.
你可能更容易从右尾部那群人里挖到人才。
You you have a better shot maybe at getting some people out of the right hand tail.
对吧？
Right?
高回报的那批人，随着池子扩大，你反正总会筛掉失败者，所以量就特别关键。
The high return, highest return people, as you expand the pool, you're gonna have You're always gonna lop off the losers anyway, so volume is gonna be very important.
是的。
Yeah.
纯粹靠数量就能占优势。
Just pure numbers is gonna be a good thing.
你看现在最火的两家初创公司，Facebook 和 Google，已经说明问题了。
And, you know, the two You see it already, like the two hottest start ups right now, Facebook and Google.
对吧？
Right?
你还把谷歌算作初创公司吗？
If you still count Google as a startup.
它们刚起步的时候风险都特别大，风险投资机构根本不愿意碰。
And both of them were so risky when they first started that venture capitalists wouldn't hear of it.
两家公司最初都只能找天使投资人筹资，因为风投都拒绝了它们。
They both initially had to raise money from angels because VCs told them both no.
你解释一下风投和天使投资的区别吧。
Explain the distinction between VCs and angels.
风投是拿别人的钱投大笔资金，天使投资人则是用自己的钱投小笔资金。
Oh, VCs, invest large amounts of other people's money, and angels invest smaller amounts of their own money.
是的。
Yeah.
对吧？
Right?
比如罗恩·康威。
So Ron Conway.
罗恩·康威是一位著名的天使投资人。
Ron Conway is a famous angel.
事实上，他个人投资了谷歌和Facebook两家公司。
In fact, he was He didn't He angel invested in both Google and Facebook.
那为什么天使投资人反而更愿意承担风险呢？一般人可能会觉得，用别人的钱投资，应该更敢冒险，而不是用自己钱的时候。
Now, why would it be that angels I think the common guess would be that when you're investing other people's money, you could be riskier rather than when you're investing your own
不是这样的。
Oh, no.
恰恰相反。
It's the other way around.
对。
Right.
那你能解释一下为什么吗？
So explain why that is.
风险投资机构得考虑很多事。
VCs have to worry.
如果他们投了一个很烂但看起来挺靠谱的初创公司，比如创始人是那种有名气、年纪大、看着稳重的人，你知道的。
If they invest if they invest in a startup that sucks but looks plausible, like the guys are famous and old and responsible looking, you know.
一旦公司倒闭，他们就能跟投资方说：你看，这项目看起来多靠谱啊。
Then if the startup tanks, they can say to the people whose money they were investing, but look, it looked so solid.
其他人呢，这些创始人明显就是靠名气撑着的。
Everybody else, you know, these guys were evident.
是的。
Yeah.
他们还有一大堆正面报道。
They had all this good press.
对吧？
Right?
但如果他们投了个19岁年轻人，搞了个荒唐的点子，最后失败了，所有有限合伙人肯定都会问：你怎么会投给这么个年轻人，还信他这种蠢主意？
Whereas if they invest in some 19 year old with some wacko idea and it tanks, all their limited partners, they'll be saying, how could you invest in this 19 year old with this stupid idea?
到时候他们又能怎么解释呢？
And then what are they gonna say?
对吧？
Right?
而天使投资人呢，完全凭直觉做决定。
Whereas an angel, an angel can just let go with his gut.
对吧？
Right?
而且，天使投资人通常属于另一类人。
Also, angels tend to be a different kind of person.
他们大多是自己曾经创办过初创公司的。
Angels are very in people who were startup founders themselves.
他们就是靠那笔创业赚的钱来做天使投资的。
That was where they got the money to do angel investing.
对吧？
Right?
而风险投资公司的人，很少有是初创公司创始人出身的。
Whereas, VCs rarely are startup founders.
他们是通过MBA这类平行路径成长起来的人。
They're people who've come up through this parallel course of MBAs and stuff.
一条平行的路径。
This parallel track.
有少数人是，他们往往能成为很出色的风投，但大多数人不是。
A few of them are, and they tend to make really good VCs, but most aren't.
大多数人只是做资金管理的。
Most are money managers.
你刚才想说，谷歌和Facebook最初都是天使投资的项目，对吧？
So you were gonna say that the two hottest Google and Facebook were both angel investments.
一开始确实是。
Initially, yeah.
没错。
Yeah.
风投机构当时都拒绝了它们。
VCs Because those turn them down.
创始人并不是那些资历深厚、履历光鲜的中年人。
Founders were not old guys with long track records and resumes.
这个想法听起来简直荒唐。
And the idea seemed wacko.
对吧？
Right?
Facebook刚成立的时候，‘社交网络’这个词根本还不存在。
Like, there was no the the word social networking hadn't been invented when when Facebook was founded.
就像托尔金写《魔戒》的时候，‘奇幻’这个类型也还没被定义出来。
Just like fantasy as a genre didn't exist when Tolkien wrote Lord of the Rings.
对吧？
Right?
那《魔戒》算什么？
So what was the Lord of the Rings?
就是个怪怪的童话故事罢了。
It's this weird like fairy tale.
那到底是什么呢？
What was it?
对吧？
Right?
任何真正新颖的想法，都会开创一个全新的领域。
Same for any really new idea, you know, creates the genre.
所以当时根本没这个名字，不管它到底是什么。
So there's no name for it, whatever it is.
社交网络这件事之所以这么有趣，部分原因就在这儿。
And the social networking thing is so fascinating partly because of that.
对吧？
Right?
它在几年前还根本不存在，现在却影响了成千上万人的生活。
It's it's something that's taken over thousands of people's lives didn't exist years ago.
是的。
Yeah.
这让人不禁想知道，接下来还会有什么出现。
It makes you wonder what else is coming.
没错。
Right.
确实如此。
Well, that's right.
这让人琢磨它到底是什么，直觉上会觉得，这种东西以后再也不会有了。
It makes you wonder what it is and the impulse says, well, there won't be anything like that ever again.
但显然，大概三年后就会有新的了。
But of course, there will be in about three years probably.
不管那是什么，现在可能已经在研发了。
It's Whatever it is, is probably being built now.
对。
Yeah.
就是你没法预知它到底会是什么。
It's just you can't tell what it is.
你知道吗？
And you know what?
我也说不准。
I can't tell either.
是的。
Yeah.
你记得吗？之前你问过我，很久以前，有没有什么模式或想法，对吧？
You I remember you asked me earlier, much earlier, if there were any patterns and ideas, right?
那时候，《圣徒》里有一句台词。
And, you know, there was a line in The Saint.
电视剧？
TV show?
对。
Yeah.
就是那部叫《圣徒》的电视剧，罗杰·摩尔主演的，其实我挺推荐这部剧的。
The TV show The Saint, where Roger Moore If you you shouldn't I would recommend The Saint, actually.
你会对罗杰·摩尔刮目相看的。
You'll have a new respect for Roger Moore.
他可不是那个差劲的邦德。
He's not just like the bad Bond.
他在《圣徒》里演得特别棒。
He was great in The Saint.
有一句台词是，别人说你对我很失望之类的。
There's a line where someone says, you know, you're you're disappointed by me or something like that.
他回答：我从来不会对任何事失望。
And he goes, I'm never disappointed by anything.
意思是他已经学会了，再没什么能让他惊讶了。
Basically, he's learned he's never surprised by anything anymore.
对吧？
Right?
我也明白了这一点，因为我们现在已经投资了144家初创公司。
And I've I've learned that too because I we've now we've now funded a 144 startups.
对吧？
Right?
就在那之前，我认识的几乎每个人都曾经创办过创业公司。
And even before that, like, practically everybody I knew was a startup founder at some point or other.
对吧？
Right?
所以我亲眼见证了太多事情的发生。
So I have seen so much stuff happen.
我多次被那些看起来前景极好却毫无建树的人失望，也多次被那些看似平平无奇却做出惊人成绩的人惊艳。
And I have been surprised so many times by people who seemed super promising and produced nothing, and other people who seemed lame and produced amazing stuff.
现在，我就只是静静地看着事情发展。
At this point, I just, like, sit back and watch things happening.
挺有意思的。
That's cool.
你知道的。
You know?
我的意思是，我确实会留意一些东西，但不是特定类型的想法。
I mean, there are things I look for, but not certain kinds of ideas.
你并不是在试图预测下一个大热门。
You're not trying to anticipate the next big thing.
不是。
No.
按定义来说，根本做不到。
By definition, you can't.
是的。
Yeah.
对吧？
Right?
按定义来说，根本做不到。
By definition, you can't.
如果真是个好点子，那它听起来往往会有点傻。
If it's a really good idea, then, you know, it'll seem kinda stupid.
这就是资本主义的厉害之处。
That's what's cool about capitalism.
不管怎么说，它就该是这样的。
It should be cool about it anyway.
对。
Yeah.
我又把你的话题带偏了。
So I pulled you off track again.
你想想，要是得靠中央计划来搞这些，会怎么样。
Just imagine if you had to do central planning for this.
没错。
Right.
我就在这儿，身处核心，但我明确地不打算去规划什么。
I've just said, here I am, I'm at the heart of this, and I have an explicit policy of not even trying to plan.
就算我想提前去琢磨该做哪类项目，也根本没用，对吧？
If I even tried to think about think ahead about idea about type types of projects anyway, right?
那只会让我的思路被一堆废话拖累。
I It would only add It would only dilute my thinking with crap.
是的。
Yeah.
你懂的吧？
You know?
你看看欧洲那边，法国和德国政府联手想搞一个自己的谷歌。
You see in in There's this story in Europe where the French and German governments are cooperating to make some kind of Google.
他们打算砸一亿美元做自己的谷歌，这简直太荒唐了。
They're gonna spend like a $100,000,000 making their own Google, and it's so stupid.
是的。
Yeah.
这想法在1997年或许还行得通。
Like, that was a good idea in 1997.
是的。
Yeah.
他们错过了机会。
They missed it.
不，根本不行。
Just No.
他们错过了机会。
They missed it.
窗口已经关上了，对吧？
Like The window closed, didn't I think?
如果他们真想搞一个法国版的谷歌，就不该是个搜索引擎。
Well, if they wanna have a Google of France, it shouldn't be a search engine.
对吧？
Right?
应该做点别的东西。
It should be something else.
那该做什么呢？
What should it be?
那咱们来想想看。
Well, let's think.
谷歌当初是什么？
What was Google?
谷歌刚出来的时候，大多数人觉得这主意蠢透了，还是几个外行提出来的。
What was what Google was was something that at the time looked like a stupid idea to most people proposed by a bunch of outsiders.
对吧？
Right?
政府项目怎么可能搞成这种外行提出的蠢主意呢？你想想看。
What are the odds that some government project is gonna end up working on a stupid idea of the type proposed by outsiders, you know?
成功率比平均水平还低，比瞎蒙还差。
Worse than average, worse than random.
没错。
Sure.
不可能。
No.
这正是哈耶克所说的试错机制的绝佳体现，进步正是来自这种过程。
It's a beautiful example of, you know, this the Hayekian idea of trial and error is where progress comes from.
它不是自上而下规划出来的。
It's not centrally planned.
而是由一个个个体推动的，很多人会失败，市场会淘汰他们，然后我们才能向前走。
It's planned by individuals who many of whom fail and the market weeds them out and we move forward.
事实上，我们建议初创公司运作的方式，就是各自独立地去尝试。
Well, fact, the way we advise startups to operate is to do this individually on a
每个创业项目都这么干。
per startup basis too.
我们给的建议也没什么特别的秘诀。
What we advise and it's not like some magic secret of ours.
这在网页应用领域其实是标准做法。
This is pretty much standard procedure with web apps.
我们会告诉人们：能上线就赶紧上线。
We tell people, release something as soon as you possibly can.
因为发布产品的目的，是为了从用户那里了解你真正该打造的是什么。
Because the point of releasing is to start learning from your users what you should have actually been building.
对吧？
Right?
是的。
Yeah.
所以，只要你做出点像样的东西，哪怕只有十个人会感兴趣，就赶紧发布出去。
So you release something as soon as you have anything that's, like, good enough that even 10 people would care.
然后仔细观察这十个人是怎么用的，这时候你才真正开始打磨你的产品，不管它是什么。
And then you watch really carefully what those 10 people do and that's when you start actually building your product, whatever it is.
是的。
Yeah.
我们之前和马比·戴谈过，说在美国经济的很多地方，信息其实是自下而上流动的。
Know we did an we had a conversation with a Marby Day about how information goes up rather than down in the American economy, in lots of lots of places.
产品创造者是从用户那里学到东西的。
The creators of the products learn from the users.
这是一种
It's a
非常棒的方式，再没有比拥有那些有见识的用户更好的了。
wonderful There's form of R and nothing better than to have sort of sophisticated users.
对吧？
Right?
千万别去为大型企业开发产品，更别提国防部了，那样的话，客户的愚蠢会把你拖得也变蠢。
And and and it's dangerous to build a product for giant corporations or God help you for the defense department because then the stupidity of your customer will, like, make you stupid.
会影响你，没错。
Affect you, yeah.
对吧？
Right?
你应该开发一款产品。
You wanna build a product.
要为那些聪明但资源有限的局外人开发产品。
You wanna build a product for clever impoverished outsiders.
没错。
Right.
对吧？
Right?
我是希望，外面有不少人和他们品味相似，足够支撑销售。
If I'm hoping there's enough people somewhat like them with their tastes out there to sell if
不是的。
you No.
关键是，那些聪明又贫穷的局外人，其实是引领者。
The point is the clever impoverished outsiders are leaders.
他们是先兆，确实如此。
They're harbingers Yeah.
就像中世纪的象征那样。
You know, in the medieval symbol.
在科技领域，几乎从定义上就是这样。
Almost by definition when it comes to technology.
对吧？
Right?
他们既精明又便宜。
They're sophisticated and they're cheap.
所以他们愿意寻找更新、更便宜也更好的东西，也乐意换新的。
So they're willing They're looking for something new and less expensive and better, and they're willing to switch.
对吧？
Right?
他们不像大公司那样被惯性驱动。
They're not just driven by inertia like large corporations.
明白了。
Okay.
那你把这些人都拉进来。
So you bring these folks in.
好的。
Okay.
好的。
Okay.
所以我们把这些创业者请进来，接下来三个月我们会全力协助他们，打造出一个能向下一波投资人展示的原型。
So we bring these folks in, and then for three months, we work intensively with them to get something that they can demo to the next ground of investors.
我们把自己定位成第一档。
We describe ourselves as like first gear.
风险投资其实通常分好几个阶段，就像变速箱的档位一样。
Venture venture funding actually tends to operate in sort of several separate levels like gears.
对吧？
Right?
很少有人能同时参与好几个阶段。
And and it's rare for people to participate in multiple levels.
有。
There's
所以这个有点像孵化器那种模式
So this is the incubator ish kind
这种说法其实挺常见的
of That's word people
嗯，这就像有个胚胎或者胎儿一样，会慢慢成长，而我之所以不
like Well, to it has a certain there's an embryo, fetus, whatever.
我之所以不这么认为是因为
It's gonna grow and Well, the reason I don't
‘孵化器’这个词之所以不贴切，是因为在互联网泡沫时期，这个词刚流行的时候，意思是所有初创公司都得在你们的场所里办公。
the reason the word incubator doesn't apply is because what it always used to mean back in the bubble, in the Internet bubble when this term first started to be used is that all the startups work on your premises.
对吧？
Right?
在你们的
Under your You
有一栋大楼，所有初创公司都在里面办公。
have some big building and all the startups work there.
我觉得这其实是个糟糕的主意。
And I think that's actually a terrible idea.
这让初创公司
It makes the startups
反而是好事。
a good thing.
而且会让初创公司的人觉得自己像雇员。
Well, also, makes the startups feel like employees.
对吧？
Right?
这可不好。
Which is bad.
是的。
Yeah.
你想想那些特别成功的初创公司，几乎都有这样的故事：刚起步时在车库里办公，那会儿是什么样子。
Very often, if you think of very successful startups, they all have these stories about what things were like in their first office when they were working out of the garage Right.
或者像那个搞浪漫的家伙那样。
Or when they, like romance guy.
他们常常熬到凌晨两点，周围全是披萨盒，就在车库里，天冷了就开洗衣机取暖，诸如此类的破事儿。
They were working until 02:00 in the morning and surrounded by pizza boxes, and they were, like, you know, in this garage with this washing machine that they would turn on to keep warm when it got cold or some crap like that.
对吧？
Right?
这可不是偶然。
It's not a coincidence.
初创公司自己单独干反而更好。
It's good for the startup to be off on their own.
他们能找到地方。
They can find space.
可以在家里办公。
They can work out of their apartments.
没错。
Right.
地方根本不是问题。
Space is not the hard part.
对。
Right.
对吧？
Right?
所以我们不这么干，就这样。
So we don't we don't do that Okay.
明明白白的。
Explicitly.
但这三个月里，他们确实得离得近一点。
So so but for those three months, but they do have to be a little bit nearby.
嗯，这正是
Well, this is what
我们做的。
we do.
我们每周一起吃一次晚饭。
We have dinner once a week.
好的。
Okay.
他们不住我们这儿，但每周会过来吃晚饭。
They don't live with us, but they come over for dinner.
不是这样的。
Not right.
所以他们每隔一段时间就会
So once So they get
来用洗衣机。
to use the washing machine.
年轻的时候，午餐特别重要。
Lunch is very important when you're young.
说实话，我觉得我们并没有。
You know, I don't think we have Okay.
但人们确实在不同时间，用各种方式大量消耗我们的基础设施。
But people do use bizarre bizarrely large amounts of our infrastructure for various things at various times.
确实如此。
It's true.
好吧。
Okay.
我们每周都会办一次晚餐，邀请一位创业领域的专家来分享，通常是某位知名的创业创始人。
So we have these dinners once a week, and we invite some expert in startups to speak, usually some famous startup founder.
对吧？
Right?
这些嘉宾会讲一些真正运作起来是怎么回事。
And and these guys give a talk about, like, how things really work.
我跟你说，我了解的关于硅谷的那些有趣八卦，大概有六成都是从这些分享会上听来的，因为全程不对外公开，他们都是创始人之间掏心窝子地聊，真的会讲干货，特别棒。
These talks, I swear, like, maybe 60% of the interesting gossip that I know about Silicon Valley, I have learned in these talks, because they're completely off the record, and these these guys, like, they're speaking founder to founder, and they really give you the goods That's cool.
讲的是事情背后的真相。
On how things really work.
这些分享真的特别有意思。
They're they're just fascinating.
可惜的是，这些内容永远没法公开发布。
Unfortunately, we can never publish them.
对。
Right.
要是真播出去了，他们立马就不说了。
I mean, if we ever broadcast them, they would just disappear.
对。
Right.
是吧？
Right?
讲者们都会闭口不谈。
The speakers would clam up.
不过这些分享真的特别有意思。
But, boy, the talks are very interesting.
好吧。
Okay.
所以，大概十周后，我们会举办一个大型活动，叫演示日，不过这名字现在有点名不副实了，因为实际持续两天。
So then, at the end of, roughly ten weeks, we have this big event called demo day, which is now a misnomer because it's two days.
我们现在的场地已经装不下所有人了。
We can't fit everybody in our building all at once anymore.
我们还是会办这个大型演示日，邀请硅谷和全美各地的投资人来参加。
We have this big event though called Demo Day where, we invite all the investors in Silicon Valley and the rest of the country.
他们从各地赶来，每家初创公司都有五分钟时间做展示。
They come in from some other places, and the startups all present to them for five minutes apiece.
对吧？
Right?
就这么简单。
That's it.
一切都很快。
Everything's fast.
一切都高效得很。
Everything's efficient.
然后啪啪啪，26家初创公司一一亮相，接下来就是关于如何融资下一轮的讨论了。
And, you know, bang bang bang, all 26 startups present, and then that's sort of the beginning of the conversation about raising the next round of money.
三个月结束后，聚餐就没了，但其他一切照旧。
At the end of three months, the dinners stop, but nothing else stops.
我每周都有固定的办公时间。
I have regular office hours.
我会跟这些初创公司一一见面。
I meet with the startups.
就连我们两三年前投资过的公司，我现在还在跟他们开办公会。
And, like, I'll I'm still having office hours with startups we funded two, three years ago.
所以，三个月后唯一正式结束的，就是聚餐这一项。
So there's the only formal thing that ends after three months is the dinners.
所以，这种正式的辅导和指导环节，对很多其他项目来说可能是缺失的。
So it's a there's a formal counseling and mentoring experience that's part of this that might be missing for a lot
其他的就没有了吗？
of other or no?
是的，没错。
Oh, yeah.
对。
Yeah.
这是很重要的一部分，但这一点真的很棒。
So it's an important part, but that's a very cool thing.
不是这样的。
Oh, no.
主要就是这个意思。
That's what it's mainly about.
不是为了钱。
It's not about the money.
没错。
Right.
不是为了钱。
It's not about the money.
关键是知识。
It's about the knowledge.
对。
Yeah.
你会帮他们谈生意吗？
Do you help them with the deal making?
会的。
Yes.
这是个大问题。
A huge problem.
我们帮初创公司主要做两件事：一是产品开发，二是跟投资人沟通。
The two big things that we work with startups on are what they're building, and talking to investors.
这两件事最重要。
Those are the biggest things.
另外还有一些次要的事情，比如
And then there's a bunch of other peripheral things like
但他们不知道该怎么跟投资人沟通。
But they don't know how to talk to investors.
大多数人
Most of
我们？不会。
us No.
没错。
Right.
我们得教他们如何
We have to teach them how to
谈判并设计下一轮融资结构，因为这非常关键。
negotiate and structure that next deal because that's a huge
谈判其实并不难。
The negotiation part isn't the hard part.
要是真那么简单就好了。
You wish.
对吧？
Right?
最难的是让投资者愿意坐下来谈。
The hard part is to get an investor to wanna even negotiate.
是的。
Yeah.
没错。
That's true.
是的。
Yeah.
最难的是让投资者相信你的初创公司值得投资。
The hard part is to convince the investors that your startup is worth investing in.
对吧？
Right?
所以我们主要跟他们聊的就是这一点。
And so that's the main thing we talked to them about.
那你们跟他们谈谈判细节的时候，都聊些什么呢？
And what's your talked to them about the details of negotiation too.
你们占多少股份？
And what's your share?
从这个经历里，你们能得到什么？
What do you guys get out of this experience?
股份。
Equity.
比例不一样，有的多有的少，每次谈判都不一样，具体情况看情况定。
Same amount, different amounts, negotiation, deal by deal, how do you It varies
大概在百分之二到百分之十之间。
from two to 10%.
不过平均下来，通常是百分之六到七。
But on average six to seven.
明白了。
Okay.
还有普通股
And The common stock
就像创始人一样，在后续轮次中会被稀释。
and being diluted in later rounds just like the founders do.
大家对这个怎么看？
What do people think of this?
Y Combinator这个理念？
The y combinator idea?
你的声誉怎么样？
What's your what's your reputation?
我觉得还不错。
I think it's good.
我可以给你一个证据，说明我们口碑很好：当有人申请我们的时候，我敢肯定地告诉他们，随便挑一家Y Combinator投资的公司去问，他们都会说我们不错。
And I'll tell you one evidence, one sign that I have that it's good is that when, when people are applying to us, you know, I feel confident telling them that they can just pick any y-combinator funded startup and ask them what they think of us.
对吧？
Right?
如果我觉得有不少人不喜欢我们，我就会对这么做感到不安。
If our if I thought people if they if I thought there were a significant number of people who didn't like us, I would feel nervous about doing that.
是的。
Yeah.
所以要么我们的口碑不错，要么就是我被稀释了。
So either our reputation is good or I'm diluted.
对。
Yeah.
我觉得口碑不错。
But I think it's good.
我觉得口碑不错。
I think it's good.
没错。
Uh-huh.
如果口碑不好，我会吓坏的，因为在这种行业里，声誉就是一切。
I would would be horrified if it wasn't because in this in this kind of business, your reputation is everything.
那我们来聊聊我们所处的这个时代吧。
So let's let's turn to the times we live in.
有两个问题。
Two issues.
作为这个领域里深度参与的一员，你有没有注意到，自去年12月7日以来的宏观经济衰退，特别是加州面临的财政问题，带来了什么影响？
As a player in this a deeply embedded player in this world, what have you noticed to be the impact of the macroeconomic recession going back to December 7 and then California in particular has having some problems Yeah.
财政方面。
Fiscally.
你有没有察觉到，这些情况对你们Y Combinator的人有什么实际影响？
Have you noticed, has it affected your life as a y-combinator person in terms of anything?
加州的财政困难，我根本没感觉到任何迹象。
California's fiscal troubles, I have not noticed any sign of.
这似乎完全没造成什么影响。
That doesn't seem to affect anything at all.
也许以后会有吧。
Maybe it will.
是的
Yeah.
也许会吧。
Maybe it will.
如果输水渠断流了。
If the aqueducts stop flowing.
他们说的是对的吗？
Are they Right?
这里的税很高，或者住这儿会没那么舒服了。
They taxes a lot or might be less fun to live here.
对。
Yeah.
那可能会影响一些事情。
That could affect things.
但这里的税已经高得离谱了。
But taxes here are already hideously high.
没错。
Right.
所以，想想他们现在收的所得税已经有多高了。
So, think what it's Income gonna tax already.
我不敢想象如果再涨上去会变成什么样。
I don't think what it's gonna be like when it's higher still.
这可能会带来一些影响，比如
It could have some effects that would
确实如此。
be It's true.
最让人担心的是，他们能把税率提到多高，才会让人们开始离开硅谷。
The scary thing It's scary to think how high they could raise taxes before people would start to leave Silicon Valley though.
他们还是很喜欢这里。
They like it a lot.
我不希望政府知道这一点。
I don't want the government to know that.
是的
Yeah.
不好意思。
I'm sorry.
那整体经济呢？
What about the
整体经济？
economy at large?
整体经济？
The economy at large.
我们其实挺担心的。
The We were worried, actually.
我们非常担心投资者会收紧钱包，不再投资了。
We were very worried that that in the investors would slap their wallets shut and nobody would be investing.
但人们还是继续投资。
But people kept investing.
我说‘保持’用过去式，是因为至少对初创公司来说，最难的日子已经过去了。
And I say kept in the past tense because for for startups, at least, the bad days are over.
对初创公司而言，除非再出现第二波灾难，比如中国股市科技股那种情况——但愿别这样——否则至少初创公司的低谷已经过去了。
For startups, unless there's some second wave of disaster, like the Chinese stock market techs or something like that, knock on wood, unless there's some new wave of disaster, for start ups at least, the the the badness has bottomed.
当时情况最糟的时候，你注意到了什么？
When it was bad, what did you notice?
估值降了。
Valuations were lower.
大家还是在投资，但估值已经跌到以前的一半了。
People were still investing, but valuations were like half what they had been.
也就是说，有哪些机会呢？
Meaning What opportunities
是什么？
would be?
不是。
No.
我的意思是，当投资人投资初创公司时，估值是怎么算的。
I mean, valuations when people invest in a startup.
如果投资人投了一百万美元，而投前估值是两百万，那投后估值就是三百万，也就是把这一百万加进去。
If people invest in a startup, if they invest like a million dollars at a pre money valuation of 2,000,000, what that means is the post money valuation is 2,000,000 plus the million they put in or 3,000,000.
所以他们就拿到了公司三分之一的股份。
So they're gonna own a third of the company.
对吧？
Right?
所以我说估值低，其实是说投资人要求拿走更多的股份。
So effectively, when I say valuations were lower, what I mean is investors were asking for more of the company.
他们当时很担心。
They were scared.
说实话，当时对未来充满了不确定性。
You know, I A lot uncertainty about the future.
但我真不觉得投资人是害怕的。
I honestly don't think the investors were scared.
我觉得投资者只是感觉到大家都更急着融资，所以这纯粹是供需关系的问题。
I think they just that investors just sensed that everyone was more desperate, and so people so it was just supply and demand.
有些人被经济形势打乱了节奏，投资的人变少了，而还在投资的人就想，好吧，我们可以把条件提得更苛刻一点。
There were some people who were sort of thrown off by the economy, and so there were fewer people investing, and the people who were still investing thought, alright, we can Well, they could.
我们可以尽量在条款上占便宜。
We can push things on terms.
不对。
No.
没错。
That's right.
能捞多少算多少。
We'll get away with what we can.
是的。
Yeah.
他们一向都是这么做的。
Well, they always try.
这关乎竞争，竞争少了，他们就能拿到更好的条件。
It's a question of competition, and when there's less competition, they're gonna get a better deal.
没错。
Yeah.
但并不是所有人都没钱了。
But so that Not not everybody ran out of money.
对吧？
Right?
有些风投基金确实出了问题，因为他们的有限合伙人没法按时出资，结果根本没法继续投资了。
There were some VC funds that were, like, really sort of structurally messed up because their LPs couldn't come up with their capital commitments and stuff like that, and they literally couldn't keep writing checks.
对吧？
Right?
有限合伙人就是LP。
LPs are limited partners.
是的。
Yeah.
但还有很多其他投资者，比如大型天使投资人和顶级的风投基金。
But there were plenty of other investors like big angels and the more, you know, the top VC funds.
他们根本不受影响。
They had no problem.
他们一直在持续投资。
They kept investing.
所以他们趁经济不景气，拿下了更多人的股份。
So they just basically took advantage of the bad economy to get more of everyone's You'd
你可能会觉得，有些失业的人会转去创业，作为下一步的打算，而不是原本计划要做的事。
think there'd be some people who were thrown out of work who might turn to a startup as a way to do the next thing they were might have done otherwise, instead the thing they were gonna have done otherwise.
我们没看到太多这样的情况。
We didn't see a lot of evidence of that.
我不能肯定说完全没发生过，但确实没见到多少。
I wouldn't say for sure that it hadn't happened, but we haven't seen a lot of it.
你们通常合作的人年龄多大？
What's the age typically of the people that you're working with?
大概二十五六岁，刚毕业没几年吧。
How old are 25, 26, you know, a couple years out of college.
他们要么在大公司干了几年，觉得这日子真没劲。
So, either they've been working for some big company for a couple years and realize, you know, oh, this sucks.
对。
Yeah.
或者正在读研，干脆退学了，类似的情况。
Or, they're in grad school and drop out or something like that.
我们之前聊过城市的事，你最近一篇文章也提到，城市能不能抓住创业市场，成为吸引初创公司的磁石。
We were talking earlier about how cities, and you wrote about this in a recent essay, how cities might be able to capture the startup market or attract, be a magnet for startups and
风险投资
venture I
尤其是九十年代的城市经济发展，曾经走过一段挺奇怪的路，大家都觉得咱们得学西雅图。
think economic development in the nineties in particular went through, urban economic development went through a rather bizarre phase where people thought we need to be more like Seattle.
这
And it's a
说实话，对大多数美国城市来说，这倒是件好事。
shame Well, for to mention most US cities, that would be a good thing.
没错。
Right.
话是这么说，但问题在于，从现状到那个目标，根本走不通。
It would be, except they don't you can't get there from here is the problem.
比如，你刚才提到了圣
So for example, you mentioned St.
路易斯之前提到过，把那里作为创办一个能跟Y Combinator竞争的机构的起点。
Louis earlier as possible place to start a competitor for the y-combinator.
我以前在圣路易斯华盛顿大学待过。
I used to be at Washington University in St.
圣路易斯这座城市当时觉得，
Louis, the city of St.
我们跟西雅图挺像的。
Louis thought, well, we're a lot like Seattle.
他们有一所世界一流的大学，就是华盛顿大学。
They have a world class university, which was University of Washington.
我们也有，圣路易斯大学也是世界一流的。
We have a world class university, University in St.
路易斯。
Louis.
他们有波音，我们也有波音。
They have Boeing, we have Boeing.
人家西雅图有波音，我们圣路易斯有麦道公司。
They have Boeing, we have McDonald Douglas.
后来两家公司虽然没合并成一家，但西雅图还有微软这样的创业公司，可圣路易斯呢
Now that wasn't part of the combination, but they had an entrepreneurial company called Microsoft and St.
圣路易斯以前有孟山都，那是一家生物科技领域的健康类公司。
Louis had Monsanto, which was a biotech health type company.
我们本来也能打造出类似的产业协同效应，但有个问题：大多数人更愿意住在西雅图，而不是圣路易斯。
And we could create the same kind of synergies except for one thing, more people wanna live in Seattle than St.
关于 Bayt 播客
Bayt 提供中文+原文双语音频和字幕，帮助你打破语言障碍，轻松听懂全球优质播客。